View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by OPUS: Open Uleth Scholarship - University of Lethbridge Research Repository
University of Lethbridge Research Repository OPUS http://opus.uleth.ca
Theses Arts and Science, Faculty of
2006 Phytoremediation of pharmaceuticals with salix exigua
Franks, Carmen G.
Lethbridge, Alta. : University of Lethbridge, Faculty of Arts and Science, 2006
http://hdl.handle.net/10133/536 Downloaded from University of Lethbridge Research Repository, OPUS
PHYTOREMEDIATION OF PHARMACEUTICALS WITH SALIX EXIGUA
CARMEN FRANKS Bachelor of Science, University of Lethbridge, 2004
A Thesis Submitted to the School of Graduate Studies of the University of Lethbridge in Partial Fulfilment of the Requirements for the Degree
MASTER OF SCIENCE
Department of Biological Sciences University of Lethbridge LETHBRIDGE, ALBERTA, CANADA
© Carmen G. Franks, 2006
PHYTOREMEDIATION OF PHARMACEUTICALS WITH SALIX EXIGUA
CARMEN G. FRANKS
Approved: September 28, 2006
Dr. Stewart B. Rood, Supervisor, Department of Biological Sciences
Dr. Alice Hontela, Thesis Committee Member, Department of Biological Sciences
Dr. Bryan E. Kolb, Thesis Committee Member, Department of Neuroscience - Cdn Ctr for Behavioural Neuroscience (CCBN)
Dr. David M. Reid, External Examiner, University of Calgary, Department of Biological Sciences
Dr. Mathew G. Letts, Chair, Thesis Examination Committee, Department of Geography
ii Abstract
Municipal treated wastewater entering rivers contain biologically active pharmaceuticals capable of inducing effects in aquatic life. Phytoremediation of three of these pharmaceuticals and an herbicide was investigated using Sandbar willow (Salix exigua) and Arabidopsis thaliana. Both plants were effective at removing compounds from solution, with removal of 86% of the synthetic estrogen, 17α-ethynylestradiol, 65% of the anti-hypertensive, diltiazem, 60% of the anti-convulsant, diazepam (Valium®), and
51% of the herbicide atrazine, in 24 hours. Distribution of compounds within roots and shoots, in soluble and bound forms, differed among compounds. Uptake and distribution of pharmaceuticals within the study plants confirmed pharmaceutical behaviour can be predicted based on a physiochemical property, their octanol-water partitioning coefficients.
An effective method for detection of 17α-ethynylestradiol within surface water using solid phase extraction and gas chromatography-mass spectrometry was developed.
Previously unreported breakdown of 17α-ethynylestradiol into another common estrogen, estrone, during preparative steps and gas chromatography was resolved.
iii Preface - Thesis Structure
This research-based MSc thesis includes an introductory chapter, three stand alone
research chapters, and an integrative conclusion chapter.
Chapter 1, ‘Introduction’, provides background information on phytoremediation and the
emerging issue of pharmaceuticals entering the water environment, as well as an
introduction to the contents of this thesis.
Chapter 2, ‘Phytoremediation of trace pharmaceuticals, diltiazem, diazepam and 17α-
ethynylestradiol with sandbar willow (Salix exigua)’, represents a stand alone research
chapter. It describes the application of an abundant riparian willow species for the
removal of trace pharmaceuticals from solution. The herbicide atrazine is included as a
positive control, as it is already known to be taken up by plants.
Chapter 3, ‘Phytoremediation of trace pharmaceuticals, diltiazem, diazepam and 17α-
ethynylestradiol with Arabidopsis thaliana’, is another stand alone research chapter. The model plant Arabidopsis is investigated not for its prospective field application, but for its possible future genetic inquiry.
Chapter 4, A ‘Method for detection of 17α-ethynylestradiol in surface water,’ is a
methodology chapter. A method is outlined for extracting and measuring the synthetic
birth control hormone ethynylestradiol from waste or surface water.
iv
Chapter 5, ‘Discussion,’ integrates the information from the three prior research chapters.
Finally, Appendix A presents the statistical analyses conducted for the thesis, while
Appendix B provides additional details of the experimental methods that are not provided in Chapters 2 or 3. Appendix C provides a figure of the nuclear magnetic resonance
(NMR) analysis performed on 17α-ethynylestradiol to verify its purity. Further information on the pharmaceuticals and the herbicide used in this study are provided in
Appendix D.
v Acknowledgements
My thanks to Stewart Rood, for providing me the tools, equipment and a corner desk within his ‘sweat shop’ to do this MSc project, and to my committee members for their support. I gratefully thank David Pearce for his brain, his seemingly endless patience and for ‘three turns widdershins’.
A special thanks to David Reid for loan of the oxidizer that helped me to complete this project and Heather Bird for helping me get the oxidizer up and running.
And to Shannon MacLeod, I am eternally grateful, for you helped me through this time in
my life and humored me convincingly when I talked about this ‘stuff’.
vi Table of Contents
Signature Page ...... ii
Abstract...... iii
Preface - Thesis Structure ...... iv
Acknowledgements...... vi
Table of Contents...... vii
List of Tables ...... x
List of Figures...... xii
Abbreviations...... xv
CHAPTER 1 Introduction...... 1
1.1 Background: Pharmaceuticals in the water environment...... 1
1.2 Study pharmaceuticals ...... 4
1.3 Introduction to phytoremediation ...... 7
1.4 Study plants: Salix exigua & Arabidopsis thaliana...... 11
1.5 Plant physiology: root uptake and translocation...... 13
1.6 MSc Project summary...... 27
Literature Cited ...... 30
CHAPTER 2 Phytoremediation of trace levels of pharmaceuticals, diltiazem, diazepam and 17α-ethynylestradiol, and the herbicide atrazine, with sandbar willow (Salix exigua) ...... 38
2.1 Introduction...... 38
2.2 Materials and Methods...... 41
2.2.1 Chemicals...... 41
2.2.2 Plants, hydroponics and treatments...... 44
vii 2.2.3 Uptake studies...... 47
2.2.4 Root concentration factors and transpiration stream concentration factors. 50
2.2.5 Soluble fractions ...... 53
2.2.6 Bound fractions...... 55
2.3 Results...... 57
2.3.1 Uptake studies time course ...... 57
2.3.2 Distribution ...... 63
2.3.3 Root concentration factors and transpiration stream concentration factors. 66
2.4 Discussion...... 74
Literature Cited ...... 81
CHAPTER 3 Phytoremediation of trace levels of pharmaceuticals, diltiazem, diazepam and 17α-ethynylestradiol, and the herbicide atrazine, with Arabidopsis thaliana ...... 83
3.1 Introduction...... 83
3.2 Materials and Methods...... 84
3.2.1 Chemicals...... 84
3.2.2 Plants, hydroponics and treatments...... 84
3.2.3 Uptake studies...... 86
3.2.4 Soluble fractions ...... 86
3.2.5 Bound fractions...... 87
3.2.6 Root concentration factor and transpiration stream concentration factor.... 87
3.3 Results...... 89
3.3.1 Uptake studies time course ...... 89
3.3.2 Distribution ...... 95
3.3.3 Root concentration factor and transpiration stream concentration factor.. 103
viii
3.4 Discussion...... 107
Literature Cited ...... 115
CHAPTER 4 Method for detection of 17α-ethynylestradiol in surface water...... 117
4.1 Introduction...... 117
4.2 Materials and Methods...... 120
4.2.1 Chemicals...... 120
4.2.2 Trimethylsilyl derivatives ...... 123
4.2.3 Gas chromatography-mass spectrometry...... 126
4.2.4 Calibration curve...... 126
4.2.5 Preparative purification of Oldman River water...... 128
4.3 Results...... 132
4.4 Discussion...... 148
Literature Cited ...... 152
CHAPTER 5 Discussion...... 155
5.1 Phytoremediation of pharmaceuticals and herbicide ...... 155
Literature Cited ...... 170
APPENDIX A Statistical analyses...... 172
APPENDIX B Other experimental methods...... 194
APPENDIX C NMR analysis of 17α-ethynylestradiol ...... 205
APPENDIX D The pharmaceuticals and herbicide: digging deeper...... 207
ix List of Tables
Table 2.1. Chemical properties of the pharmaceuticals 17α- ethynylestradiol, diazepam, diltiazem and the herbicide atrazine...... 43
Table 2.2. Salix exigua summary plant information from uptake study...... 61
Table 3.1. Arabidopsis summary plant information from uptake study...... 93
Table 4.1. SIM monitoring for 17α-ethynylestradiol, 17β-estradiol-d2, and estrone (TMS derivatives) characteristic ion m/z and retention times...... 127
Table 4.2. Results from different temperature and time combinations for silylation of 17α-ethynylestradiol...... 134
Table 4.3. Spiked river water percentage recovery of 17α-ethynylestradiol after each purification step...... 145
Table 5.1. Salix exigua and Arabidopsis root and shoot fresh weight and total volume transpired over the 24 hour study period...... 160
Table 5.2. Percentage final uptake of 3 pharmaceuticals and an herbicide by Salix exigua and Arabidopsis after 24 hours, with listed log Kow values...... 161
Table 5.3. Calculated and experimentally determined root concentration factor values for Salix exigua and Arabidopsis...... 166
Table 5.4. Calculated and experimentally determined transpiration stream concentration factor values for Salix exigua and Arabidopsis...... 167
Table A2.1. Salix exigua ANOVA comparison among compounds for mean total volume transpired, root, shoot, wood and total fresh weight of replicate plants...... 173
Table A2.2. Salix exigua ANOVA comparison among compounds for cumulative transpiration volume at each sampling time...... 174
Table A2.3. Dunnett C post hoc analysis for Salix exigua for Table A2.2...... 175
Table A2.4. Spearman’s rho correlation analysis for Salix exigua between uptake and either fresh weights or cumulative transpiration volumes over the 24 hour study period...... 176
x Table A2.5. Spearman’s rho correlation analysis for Salix exigua between cumulative transpiration and fresh weights across the study period...... 177
Table A2.6. ANOVA analysis root fresh weight for Salix exigua root concentration factor plants among time, excised roots and roots attached to the cutting...... 179
Table A2.7. ANOVA analysis of root uptake in experiments with shoots removed from Salix exigua between excised roots and roots attached to cutting...... 180
Table A2.8. ANOVA analysis of root uptake in experiments with shoots removed from Salix exigua among compound over time...... 181
Table A2.9. Tamhane post hoc analysis for Salix exigua for Table A2.8...... 182
Table A3.1. Arabidopsis ANOVA analysis among compounds for averaged cumulative transpiration volume, root, shoot and total fresh weight of replicate plants...... 183
Table A3.2. Arabidopsis ANOVA comparison among the different compounds of the mean transpired volume at each sampling time...... 184
Table A3.3. LSD post hoc analysis for Arabidopsis for Table A3.2...... 185
Table A3.4. Spearman’s rho correlation analysis for Arabidopsis between uptake from solution, plant fresh weight and cumulative transpired volume at each sampling time...... 186
Table A3.5. Arabidopsis Spearman’s rho correlation analysis of cumulative transpired volume at each sampling time to plant fresh weights...... 187
Table A5.1. ANOVA comparison between final percentage uptake of the 4 compounds among Arabidopsis and Salix exigua...... 188
Table A5.2. ANOVA comparison of root and shoot fresh weight for Arabidopsis and willow among compounds...... 189
Table A5.3a-d. ANOVA comparisons for cumulative transpiration and percent uptake between Salix exigua and Arabidopsis across each sampling time for the 4 individual compounds...... 190-193
xi List of Figures
Figure 1.1. Diagrammatic depiction of basic plant root physiology and two water transport pathways...... 15
Figure 1.2. Plotted empirical relationships determined for A, root uptake with root mass varying with root concentration factor values; B, barley and hybrid poplar relationship between log Kow and root concentration factor...... 23
Figure 1.3. Plotted empirical relationships determined for barley and hybrid poplar between log Kow values and transpiration stream concentration factor...... 26
Figure 2.1. Chemical structures of the three pharmaceuticals and herbicide...... 42
Figure 2.2. Hydroponic system with Salix exigua cuttings...... 46
Figure 2.3. Culture tube set up for Salix exigua...... 49
Figure 2.4. Percentage uptake from solution by Salix exigua over the study period...... 60
Figure 2.5. Cumulative transpired volumes by Salix exigua at each sampling time for the four compounds...... 62
Figure 2.6. Distribution of recovered radioactivity from oxidation of whole root, shoot and cutting and total plant recovery of Salix exigua...... 64
Figure 2.7. Distribution of recovered radioactivity among soluble and bound fractions in roots and shoots and oxidized cutting of Salix exigua...... 65
Figure 2.8. Uptake or equilibrium curves for roots of Salix exigua...... 69
Figure 2.9. Calculated and experimentally determined root concentration factor values for Salix exigua...... 70
Figure 2.10. Distribution of recovered radioactivity within the roots and cuttings of root concentration factor experiment plants...... 71
Figure 2.11. Calculated transpiration stream concentration factor values for Salix exigua...... 73
Figure 3.1. Hydroponic system for Arabidopsis seed germination and growth...... 85
xii
Figure 3.2. Percentage uptake from solution by Arabidopsis over the study period...... 92
Figure 3.3. Cumulative transpired volumes by Arabidopsis at each sampling time for the four compounds...... 94
Figure 3.4. Distribution of recovered radioactivity from oxidation of whole root and shoot and total plant recovery of Arabidopsis...... 97
Figure 3.5. Distribution of recovered radioactivity among soluble and bound fractions in roots and shoots of Arabidopsis...... 98
Figure 3.6. Arabidopsis individual replicate plant percent recovery of radioactivity for diltiazem...... 99
Figure 3.7. Arabidopsis individual replicate plant percent recovery of radioactivity for diazepam...... 100
Figure 3.8. Arabidopsis individual replicate plant percent recovery of radioactivity for 17α-ethynylestradiol...... 101
Figure 3.9. Arabidopsis individual replicate plant percent recovery of radioactivity for atrazine...... 102
Figure 3.10. Calculated root concentration factor values for Arabidopsis...... 104
Figure 3.11. Calculated transpiration stream concentration factor values for Arabidopsis...... 106
Figure 4.1. Chemical structures of 17α-ethynylestradiol and the IS 17β- estradiol-d2 and their trimethylsilylated derivatives...... 122
Figure 4.2. The chemical structure of estrone and of trimethylsilylated estrone...... 125
Figure 4.3. Schematic diagram of the solid phase extraction system for surface waters...... 130
Figure 4.4. TIC of the products of 17α-ethynylestradiol produced during the silylation process and GC-MS analysis...... 133
Figure 4.4a. Mass spectrum of TMS-estrone...... 135
Figure 4.4b. Mass spectra of mono- and di-TMS-17α-ethynylestradiol...... 136
xiii Figure 4.5. TIC of di-TMS-17α-ethynylestradiol derivatives...... 137
Figure 4.6. Single ion chromatograms from GC-SIM analysis of 10 ng di- TMS-17α-ethynylestradiol to 10 ng di-TMS-IS...... 139
Figure 4.7. TIC from GC-MS (full scan) analysis of 10 ng di-TMS-17α- ethynylestradiol to 10 ng di-TMS-IS...... 140
Figure 4.7a. Mass spectrum for the IS...... 141
Figure 4.8. GC-MS calibration curve for 17α-ethynylestradiol and the IS...... 142
Figure 4.9. GC-MS (full scan) analysis for the first spiked river water trial...... 146
Figure 4.10. GC-SIM analysis of second spiked river water trial...... 147
Figure A2.1. Scatter plots of diltiazem treated Salix exigua plants with plant shoot, root and total fresh weight for each sampling time...... 178
Figure B1.1. Uptake of 3 concentrations of 17α-ethynylestradiol from solution with Salix exigua...... 197
Figure B1.2. Distribution of recovered radioactivity from 3 Salix exigua cuttings exposed to 3 concentrations of 17α-ethynylestradiol and harvested after 8 hours...... 198
Figure B1.3. Distribution of recovered radioactivity from 3 Salix exigua cuttings exposed to one concentration of 17α-ethynylestradiol and one harvested at 8, 24 and 48 hours...... 199
Figure B1.4. Total recovered soluble radioactivity for two methods of sample preparation prior to LSC for diltiazem and diazepam Arabidopsis shoots...... 202
Figure B1.5. Effects of bleach on LSC analysis of pigmented plant extracts...... 204
Figure C1.1. NMR purity analysis of 17α-ethynylestradiol...... 206
Figure D1.1. Plant steroid structures and mammalian estrogen steroids and progesterone chemical structures...... 210
xiv Abbreviations
ANOVA = analysis of variance ATZ = atrazine BSTFA = N,O-bis(Trimethylsilyl)trifluoroacetamide
C18 = octadecyl-functionalized silica
d2-E2 = deuterated-17β-estradiol DTZ = diltiazem DZP = diazepam (Valium®) E1 = estrone E2 = estradiol EDC = endocrine disrupting compound EE2 = 17α-ethynylestradiol fr. wt. = fresh weight GC-MS = gas chromatography-mass spectrometry hr = hour IS = internal standard
Log Kow = logarithm octanol-water partitioning coefficient LSC = liquid scintillation counter MeOH = methanol RCF = root concentration factor RT = retention time SE = standard error SIM = single ion monitoring SPE = solid phase extraction TIC = total ion chromatogram TMCS = trimethylchlorosilane TMS = trimethylsilyl TSCF = transpiration stream concentration factor WWTP = wastewater treatment plant
xv Carmen G. Franks - Introduction
CHAPTER 1 Introduction
1.1 BACKGROUND: PHARMACEUTICALS IN THE WATER ENVIRONMENT
Water contaminants have been a growing concern since industrialization, as water bodies have been treated as a convenient place for disposing tailings waters, domestic and municipal sewage, industrial waste and other effluents. The developments of pharmacology have introduced a new range of water contaminants with some pharmaceuticals being chemically very stable and hence persisting and accumulating in surface waters and even potentially in ground water.
Recently, pharmaceuticals have been detected within treated wastewater, surface water, and dinking water, such as with the discovery of clofibric acid in surface and drinking water in Berlin (Germany) in 1991 (Stan and Linkerhägner, 1992, cited in Sengl and
Krezmer, 2003). Germany and Switzerland have led the investigations of pharmaceuticals in surface and wastewater in the early 1990’s (Loffler et al., 2005), with the rest of the world quickly following their lead. In 1999, researchers within the United States undertook a nationwide reconnaissance of 95 pharmaceutical and other organic wastewater contaminants in water resources, including anti-bacterial agents, hormones, personal care products, cleaners and others (Kolpin et al., 2002). Also in 1999, Ternes et al. (1999a and 1999b) reported on the occurrence of estrogens in water from municipal sewage treatment plants of Germany, Canada and Brazil, documenting detectable levels of natural and synthetic hormones. Cargouet et al., in 2004, examined and reported on endocrine disrupting compounds (EDC) in wastewater treatment plant (WWTP) influent, effluent and receiving waters in Paris, France and its suburbs. Servos et al., in 2005,
1 Carmen G. Franks - Introduction
reported the presence of estrogens in municipal WWTP influent and effluent within
Canada. Within the province of Alberta, Alberta Environment published a preliminary
report on the presence of pharmaceuticals and EDCs in major cities’ wastewater effluent and receiving waters in 2005 (Sosiak and Hebben, 2005). Scientists within the countries of China, Japan, New Zealand, Sweden, and others, have also reported on the levels of these compounds within their waters (Komori et al., 2004; Bendz et al., 2005; Hashimoto et al., 2005; Richardson et al., 2005; Sarmah et al., 2006).
Italy recently used the same techniques established for detecting and measuring
pharmaceutical levels in water to quantify cocaine use through the river Po (Zuccato et
al., 2005). Zuccato et al. (2005) estimated that 4 kg of cocaine flow down the river daily,
a level much higher than determined by census reports.
WWTPs have varying effectiveness at removing pharmaceuticals from wastewater, as
noted by the presence of these compounds in effluent, receiving waters, and even
drinking water (Kuch and Ballschmiter, 2001). Designed to be stable for increased shelf-
and biological-life, excreted pharmaceuticals are often still biologically active as
metabolism in man or animals may be incomplete or results in an altered form that can
become active again under certain conditions (Bendz et al., 2005). Environmental and
WWTP processes such as microbial degradation, oxidation, UV degradation and
partitioning to organic particulates, may help diminish the quantity of active compounds
entering the environment (Ternes et al., 1999a and 1999b; Loffler et al., 2005).
2 Carmen G. Franks - Introduction
Agro-chemical runoff is cited as another prominent source of surface water
contamination. Included in this runoff are livestock natural hormone excretions, antibiotic and drug excretions, as well as crop pesticides and fertilizers. First-order streams and riparian corridors are considered effective means of removal and sequestering these contaminants (Angier et al., 2002). This runoff, when combined with wastewater, could add significantly to surface water contamination.
3 Carmen G. Franks - Introduction
1.2 STUDY PHARMACEUTICALS
Three pharmaceuticals were chosen for this research project, including the synthetic birth
control hormone 17α-ethynylestradiol (EE2), the antihypertensive diltiazem (DTZ), and
the anticonvulsant diazepam (DZP) (Valium®). A fourth chemical, atrazine (ATZ) is a
common herbicide that is effective in killing C-3 plants (Rao, 2000). ATZ was chosen as a positive control since it is known to be readily taken up by plants (Burken and Schnoor,
1996; Cherifi et al., 2001). Further information on the pharmaceuticals and herbicide used in this study can be found in Appendix D.
These pharmaceuticals are known to be only partially metabolized in humans with a
percentage of the administered dose excreted as parent compound, lending them to entry
into the environment. Disposal of unwanted drugs down sinks and toilets also contributes to the concentration of the active compounds entering wastewater. These drugs are administered worldwide and have been reported in the wastewaters and surface waters of many heavily populated regions (Ternes, 1998; Kolpin et al., 2002; Servos et al., 2005).
The presence of atrazine within the water environment has been documented for many years and it has even been reported in rain water (Gfrerer et al., 2002).
Hormone and hormone mimics like EE2 are known as endocrine disrupting compounds
(EDC) that interact with endogenous hormone systems in vertebrates. In aquatic
ecosystems, these can influence fish and amphibians such as by inducing the production
of vitellogenin (female egg proteins) in male fish and brown frog hepatocytes, as well as
altering fish metallothionein, a metal binding protein whose levels reflect a toxic effect
4 Carmen G. Franks - Introduction
(Werner et al., 2003; Gorshkov et al., 2004; Rankouhi et al., 2005). EE2 is a potent
estrogen with even short term exposure to EE2 resulting in decreased fertility of sexually
maturing male rainbow trout (Schultz et al., 2003). Purdom et al. (1994) reported altered
vitellogenin production with concentrations of EE2 as low as 0.1 ng/ L and EE2 has been
found in WWTP samples at levels between 1.0 and 3.2 ng/ L in Paris (Cargouet et al.,
2004). Cargouet et al. (2004) suggested that EE2 appeared to resist biodegradation and
accounted for 35-50% of estimated estrogenic activity in Parisian rivers.
Diltiazem, the antihypertensive, is of concern due to its Ca-channel blocking
mechanisms. Calcium plays a very significant role in the physiology of many organisms
and if a chemical was to disrupt this role, it could be detrimental for plants and animals.
DTZ has been detected in surface water in the United States at a maximum level of 49
ng/ L and a median level of 21 ng/ L (Kolpin et al., 2002).
Diazepam, an anti-convulsant benzodiazepine substance, was originally thought to be a
solely synthetic creation until the discovery of natural benzodiazpines in the 1960s in both plants and animals (Unseld et al., 1989; Kavvadias et al., 2000). Acting on the central nervous system, DZP has the potential to act on any organism with a nervous system (Wildmann, 1988). DZP has been detected in Belgium in wastewater influent and effluent, with levels as high as 59 and 118 ng/ L in influent and less than 10 ng/ L in effluent (van der Ven et al., 2004). Ternes et al. (1998) has also reported on DZP in sewage treatment plant effluent within Germany, with levels as high as 40 ng/ L, although it was not detected in the rivers and streams.
5 Carmen G. Franks - Introduction
Atrazine is one of the most extensively used herbicides in the world. It acts on the
photosynthetic electron transport system of plants. ATZ has been associated with
endocrine disruption at low doses (Hayes et al., 2002). Hayes et al. (2002) found evidence of ATZ induced estrogen secretion and inhibited testosterone secretion, resulting in hermaphroditic and demasculinized frogs at levels between 10 ng/ L and 100 ng/ L. Developmental deformities in amphibian larvae have also been associated with
ATZ with a dose-dependent increase in deformities with increasing herbicide exposure
(Allran and Karasov, 2001).
ATZ also indirectly affects food webs and biodiversity by hindering the growth of
primary producers such as algae, macrophytes, diatoms, phytoplankton and other
microorganisms (Dewey, 1986; Tang et al., 1997; Detenbeck et al., 1996; DeNoyelles et
al., 1982). Effects on food web foundations could be significant with prolonged and
synergistic exposure to these compounds (Kolpin et al., 2002; Fent et al., 2006).
6 Carmen G. Franks - Introduction
1.3 INTRODUCTION TO PHYTOREMEDIATION
Phytoremediation is not new, but has only been mentioned in the technical literature since
about 1994 (Schnoor, 2002). As defined by Schnoor (2002), it is
… the use of vegetation for in situ treatment of contaminated soils, sediments, and water. It is applicable at sites containing organic, nutrient, or metal pollutants that can be accessed by the roots of plants and sequestered, degraded, immobilized, or metabolized in place.
Remediation with plants has been used for inorganics such as nutrients, selenium and
arsenic, metals such as lead, cadmium, nickel and zinc (Meagher, 2000, Deng et al.,
2006; Fayiga and Ma, 2006; Pendergrass and Butcher, 2006; Shen et al., 2006).
Phytoremediation has also been used for organic chemicals (Burken and Schnoor, 1998) such as pesticides (Henderson et al., 2006), polychlorinated biphenyls (Chekol et al.,
2004), petrochemicals (Kassel et al., 2002), and explosives (TNT) (Thompson et al.,
1998; Dzantor et al., 2000; Angier et al., 2002). Heavy metal clean-up has been a large area of study for phytoremediation. To date, there has been limited research into phytoremediation of pharmaceuticals from water.
In natural ecosystems, plants have been considered as the ‘green liver.’ As photosynthetic
versions of mammalian livers, plants contain enzymes and metabolic processes that
detoxify contaminants (Sandermann, 1994). Research into metabolism of plant-intended
compounds and phytoremediation has examined the potential of many plants. Crop
plants, such as barley, corn, sorghum, and vegetables, are commonly dosed with a
pesticide or herbicide prior to harvest. Determining the outcome, effects, and methods of
7 Carmen G. Franks - Introduction
‘how it works’ uncovered the vast ability of plants to survive in contaminated environments, from anthropogenic or natural sources.
Phytoremediation may involve uptake and metabolism, rhizosphere bioremediation, phytostabilization, phytoextraction, rhizofiltration, hydraulic control, phytovolatilization, vegetative caps, and constructed wetland (Schnoor, 2002). The type of technology used depends on the compound(s) involved and the location or environment. These areas are roughly defined and tend to overlap. The focus for this MSc study involved uptake and metabolism, for which the process is generalized below.
Uptake and metabolism – using metabolic capabilities of plants to metabolize compounds into less toxic or less biologically active forms upon uptake of the compound into the plant. This process has been studied extensively in poplar trees with atrazine (Burken and
Schnoor, 1997) and the explosive TNT (Thompson et al., 1998).
Vegetation that thrives in riparian environments is a natural consideration for removing contaminants from surface and ground water. Common riparian vegetation includes phreatophytic trees and shrubs of the genera Populus and Salix, poplar and willow, respectively. These plants transpire large volumes of water, with trees using between 100 to 200 L/ day (Newman, 1997). Rapid growth is important in plants used for phytoremediation, ensuring rapid and continual uptake of the contaminant within the transpiration stream and possible storage in cell walls. Transpiration stream tension (pull)
8 Carmen G. Franks - Introduction
may play an important role in the drawing of chemicals towards the root zone and their
subsequent uptake, and this favours the use of trees that use large volumes of water.
Original reports focused on nutrient and metal remediation involving plants found to
grow on or near contaminated sites or eutrophic surface waters. Aside from Populus,
other plants include Typha (cattails), Brassica (mustard family), sunflowers, and many
other genera (Lim et al., 2003; Barrera-Diaz et al., 2004; Nehnevajova et al., 2005;
Nocito et al., 2006; Quartacci, 2006)..
Crop plants resistant to herbicides uncovered the ability of resistant plants to bind the herbicide or its metabolites to cell wall components as a form of detoxification or
sequestering, preventing the herbicide from reaching its site of action (Mathew et al.,
1998). The degree of binding and detoxification varies between plant species and
individual resistant or susceptible plants, but has been found to apply to a variety of
compounds (Langebartels and Harms, 1985; Scheunert et al., 1985; Dankwardt and
Hock, 2001; Sapp et al., 2004; Weiss et al., 2004). Bound residues, not just of pesticides
but of other chemicals, create a concern as to bioavailability after digestion or decay of the plant (Sandermann, 2004). Animals that graze or browse plants with bound residues
have the potential to release the compounds (Sandermann et al., 1990 and 1992), especially ruminants with the bacteria that break down cellulose (Skidmore et al., 1998).
The formation of bound residues is primarily through processes resulting in covalent
bonds or physical encapsulation within extracellular matrices (Skidmore et al., 1998).
9 Carmen G. Franks - Introduction
Covalent bonds are most commonly formed by the reaction of electrophilic compounds
with nucleophilic sites on proteins, nucleic acids or cell wall constituents such as lignin
or cellulose. Encapsulation, or trapping of compounds, within spaces of the cell wall is
another process proposed (Skidmore et al., 1998).
The occurrence of bound residues in food products, particularly of pesticides, has led to
countries requiring Food and Drug Act regulations that set limits on the acceptable levels within imported and exported foods. In Canada, the Canadian Food Inspection Agency
monitors pesticide residue levels on some domestic and imported foods to ensure the levels fall below set residue limits.
The costs of using plants to remediate contamination of soil or water is considered to be
lower than conventional remediation technology that often involves removal and
incineration of large quantities of earth, or the use of chelating chemicals. The long-term
applicability of phytoremediation also provides environmental and aesthetic benefits and
the in situ remediation makes it a technology worth further research and application.
New research directions in phytoremediation technology include genetic engineering of
plants and examination of natural and induced mutants (Doty et al., 2000). Now that the
poplar and Arabidopsis genome have been sequenced (Cobbett and Meagher, 2002), insight into the enzymes and the genes that code them should facilitate comparisons among plants and understanding of phytoremediation (underlying) mechanics (The
Arabidopsis Genome Initiative, 2000; Sterky et al., 2004).
10 Carmen G. Franks - Introduction
1.4 STUDY PLANTS: SALIX EXIGUA & ARABIDOPSIS THALIANA
Salix exigua. There are over 350 to 400 willow species, of the family Salicaceae, that are located primarily in the Northern Hemisphere, ranging from arctic through temperate latitudes. Most of these shrubs can be found in moisture rich regions, particularly riparian zones, with their roots obtaining water from the shallow, saturated or streamside water table below (phreatophyte). Commonly pioneer plants, the willows grow quickly on newly disturbed soil or newly formed streamside bars and banks.
Salix exigua Nutt.is known commonly as coyote willow, sandbar willow, basket willow, narrowleaf willow, slender willow, riverbank willow, acequia willow, long-leaved willow gray willow, dusky willow, and sometimes pussy-willow (Stevens et al., 2000; Nellessen,
2003), although, a more complete list of 34 synonyms exists in Kartesz (1994). Several characteristics that are uncommon for willow genus define Salix exigua, including stomata located on both upper and lower leaf surfaces (amphistomatous) and the ability to spread almost entirely clonally through creeping rootstock (Nellessen, 2003). The sandbar willow is quite tolerant of saturated soils for a prolonged period of time, but will begin to suffer if this regime is maintained for too long (Amlin and Rood, 2001).
The use of sandbar willow for phytoremediation has not been investigated, but other willow have been used in phytoremediation studies for cyanide (Larsen et al., 2005;
Bushey et al., 2006 a and 2006 b), the anti-fouling agent tributylin (Trapp et al., 2004), and cadmium and copper (Kuzovkina et al., 2004).
11 Carmen G. Franks - Introduction
Arabidopsis thaliana. Related to cabbage and mustard (family Brassicaceae),
Arabidopsis thaliana is a small flowering plant used extensively in plant biology research. Arabidopsis has become the model plant due to its small, sequenced genome, small size, short life-cycle and easy manipulation. With the complete genome sequencing of Arabidopsis thaliana in 2000, identification of genetic diversity in terms of important, generalist metabolizing and detoxifying enzymes resulted in the discovery of over 270 members of cytochrome P450 monooxygenases, as well as numerous glutathione-S- transferases (The Arabidopsis Genome Initiative, 2000; Sterky et al., 2004). This reflects the immense genetic and encoded biochemical diversity and complexity found in plants.
Recognized as a tool for the discovery of genes in the role of transformation and phytoremediation, Arabidopsis has been used for phytotransformation studies. One such study examined the TNT metabolic pathways within Arabidopsis (Subramanian et al.,
2006). Heavy metal uptake and Arabidopsis genes have been more extensively studied such as for arsenic (Geng et al., 2005) and the elements Zn, Co, Cu, Pb and Mn (Yang et al., 2005).
12 Carmen G. Franks - Introduction
1.5 PLANT PHYSIOLOGY: ROOT UPTAKE AND TRANSLOCATION
The very physiology that enables plants to live (accumulate resources and water via a
functional aqueous transportation system) is the same that can result in the uptake and accumulation of anthropogenic or natural compounds and elements. Plants have existed for millions of years and evolved hundreds of metabolic enzymes to deal with environmental compounds and contaminants. These adaptations allow for the exploitation of plants for remediation.
Plant roots. Plant root structure is quite well understood with the basic structure composed of the epidermis, cortex, endodermis (with Casparian strip), and vascular tissue
containing xylem and phloem (Figure 1.1) (Taiz and Zeiger, 2002). The epidermis is the
external layer of protective cells including those that form root hairs. Below the
epidermis are loosely packed cells, the cortex. The next layer consists of a single
encircling wall of cells termed the endodermis, sealed extracellularly with the Casparian
strip. The vascular tissue is composed primarily of xylem (water transportation tissue
from roots to shoot) and phloem (tissue that transports products of photosynthesis).
Within this tissue structure there are two barriers to plant uptake: cell plasma membranes
and the Casparian strip.
Plasma membranes of root cells are the primary barrier for ions within soil water.
Membranes are made up of a bilayer of phospholipids, creating a hydrophilic exterior and
hydrophobic interior to the membrane. Depending on a molecule’s characteristics, it may
be capable of diffusing across the membrane into the root cells and travel to the xylem
13 Carmen G. Franks - Introduction
where it will be transported to the shoots. Diffusion across the membrane is primarily a function of a physiochemical property relating to the polarity of the compound, the
octanol-water partitioning coefficient, expressed as the logarithm (log Kow), a similar
property to lipophilicity (Tracy, 2004).
14 Carmen G. Franks - Introduction
Casparian strip
(extra-cellular) Xylem Apoplastic tissue movement
Cortical cell
Symplastic movement
(intra-cellular)
Root epidermis
Root hair Endodermis
Figure 1.1. Diagram of basic plant root physiology and two water transport pathways.
15 Carmen G. Franks - Introduction
The Casparian strip forms a seal around the endodermal cells and is comprised of a waxy
material, suberin, which seals the extracellular space of the endodermis, somewhat
similar to bricks and mortar. This seal limits the bulk flow of an external solution
traveling through the extra-cellular spaces of root cells from flowing into the conducting
xylem vessels to enter the transpiration stream. Bypassing this seal requires the
compounds to cross the cell plasma membranes and travel within the cells, across the
endodermis, to the xylem tissue.
There are two pathways for water and solute uptake into the root: apoplastic and
symplastic. The apoplastic pathway involves movement through cell walls, extra-cellular
air spaces, and the xylem vessels, without crossing any membranes (Taiz and Zeiger,
2002). Bulk flow moves through this space, successfully increasing the surface area for
ion uptake and water contact with cell membrane surfaces. Movement through this space is generally inhibited at the Casparian strip. Inability to pass through the Casparian strip
may result in sorption to root components such as cell wall, either in a relatively
permanently bound form, covalently bonded or possibly encapsulated, or in a soluble
form, more reversibly bound.
A proposed symplastic transport route includes entering epidermal or cortical cells by
crossing the cell membranes and traveling within the cells, bypassing the Casparian strip
and then exiting the symplast and entering the xylem vessels (part of the apoplast) into
the transpiration stream (Taiz and Zeiger, 2002). Cell to cell transport would be via
plasmodesmata, tubular structures connecting cytosol of adjacent cells, and crossing
through adjacent cell membranes.
16 Carmen G. Franks - Introduction
Aquaporins are trans-membrane channels that allow for the selective movement of water
across membranes. The presence of these transmembrane, proteinaceous channels within
plants was discovered in 1990 (Wayne and Tazawa, 1990) and although they are not fully
understood, they may function similarly to animal aquaporins as water-selective channels or relatively non-selective channels for water and other small non-electrolytes (Tyerman et al., 2002). Relative to pharmaceuticals, the aquaporins small channel size and selectivity (currently poorly understood) likely prevent compound uptake through these pores (Tyerman et al., 2002).
Active uptake pathways require energy to move a solute across a membrane against a
concentration gradient. This requires specific transport proteins, although some transport
proteins may be more generalized and may transport compounds with similar molecular
binding sites and size to the normal target compound (Buchanan et al., 2000).
Ionization, the process of dissociation of a compound into constituent ions, may also
occur depending on their pKa (the negative log of the acid dissociation constant, Ka) and
their surrounding pH. Lipophilicity and membrane solubility are related to the charge of molecules. If a molecule becomes ionized by releasing a proton (H+) its charge becomes
negative and hence more hydrophilic. Conversely, non-ionized compounds are generally
more lipophilic. The pKa of a compound determines at what pH it will release or gain a
proton and this influences solubility in soil or surface water. After a compound crosses a
17 Carmen G. Franks - Introduction
lipid membrane, the pH within the cell and cell compartments differ, potentially altering
ionization of the compound.
Charged, ionized, particles can also move across membranes at different rates than
neutral compounds due to electrochemical charges within the cells and the associated
electrochemical gradient (Taiz and Zaiger, 2002). Cell cytosol typically has a net
negative charge, therefore, if a compound becomes positively charged while in the
apoplast, it will be attracted to the negative charge within the cell and aid in its movement
across the cell membrane. At the varying physiological pHs the pharmaceuticals of study,
DZP, EE2 and the herbicide ATZ will remain primarily non-ionized, and therefore
effectively neutral. Conversely, the pharmaceutical DTZ was obtained as DTZ-
hydrochloride, lending this compound to positive ionization at physiological pH, and
therefore the electrochemical gradients may influence its movement across cell
membranes.
Uptake and transport of weak electrolytes is typically considered a more complex process
to explain mathematically. Weak electrolyte uptake typically takes into consideration pH of different compartments, compound pKa and log Kow (the logarithm of the octanol-
water partitioning coefficient), as well as membrane permeability to ions and neutral
compounds, and the potential for ion-trapping movement within the phloem (Trapp,
2004).
18 Carmen G. Franks - Introduction
Neutral compound uptake and distribution often follow a relatively simple equation when
compared to the more complex equation for weak electrolyte uptake. Neutral compound
uptake is primarily dependant on diffusion across cell membranes into the symplast, which is typically a function of a physio-chemical property of the compounds, the octanol-water partitioning coefficient (expressed as the logarithm Kow) (Shone and Wood,
1974; Briggs et al., 1982).
Octanol-water partitioning coefficient logarithm (Log Kow). This physiochemical
measure is often determined by two immiscible phases, water and an organic solvent
(typically octanol), and the addition of a concentration of the compound to be
determined. The phases are shaken and allowed to come to equilibrium. The
concentration of the compound in both phases is determined and an octanol-water
coefficient (Kow) is determined (typically expressed as the logarithm):
Kow = [compound] in octanol / [compound] in water
The preference of a compound for the lipid-like phase (octanol) is expressed as a large
Kow value, and is similar to hydrophobicity of the compound. Compounds with high Kow
values will move out of the water phase and into the membrane lipid phase. If the Kow is large, the compound may thus not pass through the membrane, but will remain within the lipid core. Substances with very low Kow values are very hydrophilic and these compounds may not enter the cell membranes at all and will thus remain within the water
of the apoplast.
19 Carmen G. Franks - Introduction
Plants. Documentation of plant uptake was provided in 1956, by Crowdy and Jones, who
established that low rates of compound translocation within plants were associated with
strong binding of the compound within the roots. The connection between a compound’s
Kow and subsequent uptake by plants, similar to drug movement within humans, was
made in 1982 by Briggs et al..
Initial stages of plant uptake involve the creation of equilibrium between the external
solution concentration and the root. The ratio of the concentration within the root to the
concentration in the external solution is termed the root concentration factor (RCF)
(Shone and Wood, 1974). This equilibrium is met due to, and at a rate set by, the log Kow of the compound and the concentration gradient. Diffusion across cell membranes can thus occur and subsequent movement into the transpiration stream and the shoots, lends to a transpiration stream concentration factor (TSCF), a reflection of the compounds diffusion rate (Briggs et al., 1982). The transpiration stream is considered the primary channel for movement of compounds from the root to the shoot. Phloem transport is generally slight, but may occur through ion-trapping movement, much as with phloem transport of sugars and other carbohydrates (Kleier, 1988; Zebrowski et al., 2004).
Root concentration factor (RCF). The root concentration factor (RCF) was introduced in
1974 by Shone and Wood as the equilibrium partitioning between soil/solution concentration and sorption root concentration:
20 Carmen G. Franks - Introduction
RCF = concentration in the root (g/g) / concentration in external solution (g/mL).
Since RCF is a partition ratio, the values will not vary with the amount of root mass or
volume used (assuming the concentration is not saturated), but the quantity of compound taken up will vary with the root mass or volume (Figure 1.2 A). Briggs et al. (1982) empirically defined the relationship between RCF and log Kow for barley roots (Figure
1.2 B):
log (RCF – 0.82) = 0.77 log Kow – 1.52.
In 1998, Burken and Schnoor, published a similar empirical relationship for hybrid poplar
roots (Figure 1.2 B):
log (RCF – 3.0) = 0.65 log Kow – 1.57.
Although equations have been developed to predict RCF values for hybrid poplar and barley, it is likely that equations from other experiments will differ as RCF values are dependent on specific factors, such as the plant species used, environmental conditions, compound characteristics and physiochemical properties in those environmental
conditions. Dietz and Schnoor (2001) discuss how roots of hybrid poplars differ in their
absorption rates depending on their environment as roots grown in the field had higher
lipid contents than roots grown in a hydroponic system. For example, Thompson et al.
(1998) examined the uptake of TNT by hybrid poplars, determining the RCF to be
21 Carmen G. Franks - Introduction
approximately 49 mL/ g versus the calculated 1.7 and 3.5 mL/ g, using Briggs et al.
(1982) and Burken and Schnoor’s (1998) equations, respectively. Conversely, Briggs et al. (1982) reported that RCF values from other experiments fit their equation relatively
well.
In dilute solutions, RCF is independent of concentration, behaving more as a partitioning
reaction (Briggs et al., 1982). RCF has been shown to increase with increasing compound
log Kow in a non-linear fashion (Briggs et al., 1982; Burken and Schnoor, 1998).
Compounds with log Kow values > 3.0 are considered to sorb strongly to roots based on
the equations developed by Briggs et al. (1982), and Burken and Schnoor (1998).
22 Carmen G. Franks - Introduction
1.0 A
0.8 g)
μ 0.6
0.4 RCF 113.8 (EE2) Root uptake ( Root uptake RCF 23.2 (DZP) 0.2 RCF 18.5 (ATZ)
0.0 0.2 0.6 1 1.4 1.8 Root mass (g)
10000 B
1000 barley
100 poplar 10
1 Root concentration factor (RCF) (RCF) factor Root concentration 0 0123456
Log octanol-water partition coefficient (Kow)
Figure 1.2. Plotted empirical relationships determined for A, root uptake with root mass varying with root concentration factor values; B, barley (Briggs et al., 1982) and hybrid poplar (Burken and Schnoor, 1998) relationship between log Kow values and root
concentration factor.
23 Carmen G. Franks - Introduction
Transpiration stream concentration factor (TSCF). The transpiration stream
concentration factor (TSCF) is defined as the concentration within the transpiration
stream (μg/ mL) relative to the concentration in external solution (μg/ mL) (Briggs et al.,
1982). The concentration within the transpiration stream is often measured as the amount
of compound within the shoot, per volume of water transpired over the average of initial
and final external solution concentrations (Briggs et al., 1982). Since transpiration is
somewhat continuous, the mean of initial and final external solution concentrations are
used (Briggs et al., 1982).
TSCF = (concentration in the shoot (μg) / volume transpired (mL)) / ((external solution
concentration (μg/mL)initial + external solution concentration (μg/mL)final )/ 2)
The TSCF versus log Kow relationship is typically a bell-shaped, or normal curve, with
the peak representing optimum uptake into the transpiration stream that is typically at a
log Kow value around 2, although this varies across plants (Figure 1.3). TSCF generally ranges from 0 to 1, with 1 implying passive uptake that follows the transpirational flow.
Values less than 1 imply the compound is less readily moved into the transpiration stream and values greater than 1 imply active uptake, such as is typical for nutrients (Orchard et
al., 2000; Dietz and Schnoor, 2001). Briggs et al. (1982) empirically defined the
relationship between TSCF and log Kow for barley roots (Figure 1.3):
2 log TSCF = 0.784 exp[-(log Kow – 1.78) / 2.44].
24 Carmen G. Franks - Introduction
In 1998, Burken and Schnoor, published a similar empirical relationship for hybrid poplar roots (Figure 1.3):
2 log TSCF = 0.756 exp[-(log Kow - 2.50) / 2.58].
Analyses of RCF and TSCF allow for insight into a compound’s distribution within a plant, the relationships between log Kow, uptake and distribution, and thus enables comparisons between compounds and plants involved in phytoremediation. Predictable, deterministic behaviours are useful for considering the fate of pharmaceuticals in the environment.
25 Carmen G. Franks - Introduction
0.9
0.6
poplar 0.3 Transpiration stream stream Transpiration barley concentration factor (TSCF)
0.0 0123456 Log octanol-water partition coefficient (Kow)
Figure 1.3. Plotted empirical relationships determined for barley (Briggs et al., 1982) and hybrid poplar (Burken and Schnoor, 1998) between log Kow values and transpiration
stream concentration factor. Optimum uptake into the transpiration stream is represented
by the peak of the curve.
26 Carmen G. Franks - Introduction
1.6 MSC PROJECT SUMMARY
This MSc project investigated the potential application of a riparian plant for the removal
of environmental levels of pharmaceuticals from water.
The primary focus was on phytoremediation of pharmaceuticals from solution using the
riparian shrub, Salix exigua. Salix exigua was chosen as it is a common riparian plant that
is able to survive within the wet conditions of a river bank. Willow species also possess
other important traits, such as rapid growth and colonization, transpiration of large
volumes of water, and extensive geographic distribution. Levels of the chosen
pharmaceuticals used for this study are equivalent to 40 ng/ L. This level could be
considered high when compared to some reported environmental levels found within
streams and rivers, although a few have reported levels equivalent to this at the high end
of detected values. When compared to wastewater influent or effluent, this concentration
may be considered a median to high value.
Initial investigations answered the question as to whether these compounds are taken up,
and at what rate they are taken up by plants, particularly willow. Willow’s ability to
remove these compounds from solution was then compared to predicted behaviours of the
compounds based on an important physiochemical property, the octanol-water partitioning coefficient (Kow, expressed as the logarithm). This examination included a
compound distribution analysis within the plant of the compounds taken up, including the
plant components of roots, wood stem, and green shoot (including leaves and green
stem). Again, the distribution was compared to predicted behaviours of the compounds
27 Carmen G. Franks - Introduction
from their log Kow. Distribution within the plant was differentiated into soluble versus
bound forms.
Secondary focus was the ability of Arabidopsis thaliana to remove these compounds from solution. A similar investigation into uptake, rates of uptake, and distribution within the roots and shoots of Arabidopsis was performed, as for willow. Investigation into
Arabidopsis is seen as less of a field application of phytoremediation than as an important source for broadening the understanding of plant uptake. A model plant, Arabidopsis can provide insight into the enzymes and genes that play important roles in the outcome of pharmaceuticals within plants. Arabidopsis as well, proves that diverse ranges of plant genera are capable of ‘mopping up’ these environmental contaminants. A second plant genus also reaffirms that pharmaceuticals behave, to a degree, according to their log Kow
values, an important tool for predicting their relationship within plants and the
environment.
Development of a method for the detection of 17α-ethynylestradiol in surface water by
gas chromatography-mass spectrometry (GC-MS) originated from the interest as to whether this potent compound was present in local river water. In developing this method
for detecting estrogens within environmental water samples it was discovered that EE2
can degrade into another commonly detected estrogen if analyzed improperly. This had
not been reported in the literature at the time of this method development. A method for
the analysis of EE2 minimizing degradation and optimizing its detection through GC-MS
was developed. No environmental samples, unspiked, were analyzed since an
28 Carmen G. Franks - Introduction investigation by Alberta Environment was published reporting no detectable levels of
EE2 within the local river.
29 Carmen G. Franks - Introduction
LITERATURE CITED
Allran, JW and Karasov, WH. 2001. Effects of atrazine on embryos, larvae, and adults of anuran amphibians. Envrion. Toxicol. Chem. 20: 769-775.
Amlin, NA and Rood, SB. 2001. Inundation of riparian willows and cottonwoods. J. Am. Water Resour. Assoc. 37: 1709-1720.
Angier, JT, McCarty, GW, Rice, CP, Bialek, K. 2002. Influence of a riparian wetland on nitrate and herbicides exported from an agricultural field. J. Agric. Food Chem. 50: 4424-4429.
Barrera-Diaz, C, Colin-Cruz, A, Urena-Nunez, F, Romero-Romo, M, Palomar-Pardave, M. 2004. Cr(VI) removal from wastewater using low cost sorbent materials: roots of Typha latifolia and ashes. Environ. Technol. 8: 907-917.
Bendz, D, Paxéus, NA, Ginn, TR, Loge, FJ. 2005. Occurrence and fate of pharmaceutically active compounds in the environment, a case study: Höje River in Sweden. J. Hazard. Mater. 122: 195-204.
Briggs, GG, Bromilow, RH, Evans, AA. 1982. Relationships between lipophilicity and root uptake and translocation of non-ionised chemicals by barley. Pestic. Sci. 13: 495-504.
Burken, JG and Schnoor, JL. 1996. Phytoremediation: Plant uptake of atrazine and role of root exudates. J. Environ. Eng. 122: 958-963.
Burken, JG and Schnoor, JL. 1997. Uptake and metabolism of atrazine by poplar trees. Environ. Sci. Technol. 31: 1399-1406.
Burken, JG and Schnoor, JL. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32: 3379-3385.
Bushey, JT, Ebbs, SD, Dzombak, DA. 2006a. Development of a plant uptake model for cyanide. Int. J. Phytorem. 8: 25-43.
Bushey, JT, Small, MJ, Dzombak, DA, Ebbs, SD. 2006b. Parameter estimation of a plant uptake model for cyanide: application to hydroponic data. Int. J. Phytorem. 8: 45- 62.
Buchanan, BB, Gruissem, W, Jones, RL. 2000. Biochemistry and molecular biology of plants. Chapter 3 Membrane Transport.Rockville, Maryland, USA: American Society of Plant Physiologists. 1367 p.
30 Carmen G. Franks - Introduction
Cargouet, M, Perdiz, D, Mouatassim-Souali, A, Tamisier-Karolak, S, Levi, Y. 2004. Assessment of river contamination by estrogenic compounds in Paris area (France). Sci. Total Environ. 324: 55-60.
Chekol, T, Vough, LR, Chaney, RL. 2004. Phytoremediation of polychlorinated biphenyl-contaminated soils: the rhizosphere effect. Environ. Int. 30: 799-804.
Cherifi, M, Raveton, M, Picciocchi, A, Ravanel, P, Tissut, M. 2001. Atrazine metabolism in corn seedlings. Plant Physiol. Biochem. 39: 665-672.
Cobbett, CS and Meagher, RB. 2002. Arabidopsis and the Genetic Potential for the Phytoremediation of Toxic Elemental and Organic Pollutants. The Arabidopsis Book, eds. C.R. Somerville and E.M. Meyerowitz, American Society of Plant Biologists, Rockville, MD, doi/10.1199/tab.0032 http://www.aspb.org/publications/Arabidopsis/ - this publication is only available as an on-line text.
Crowdy, SH and Jones, DR. 1956. The translocation of sulphonamides in higher plants I. uptake and translocation in broad beans. J. Exp. Bot. 7: 335-346.
Dakora, FD and Phillips, DA. 2002. Root exudates as mediators of mineral acquisition in low-nutrient environments. Plant and Soil 245: 35-47.
Dankwardt, A and Hock, B. 2001. Immunolocalization of non-extractable (bound) residues of pesticides and industrial contaminants in plants and soil. Chemosphere 45: 523-533.
Deng, H, Ye, ZH, Wong, MH. 2006. Lead and zinc accumulation and tolerance in populations of six wetland plants. Environ. Pollut. 141: 69-80.
DeNoyelles, F, Kettle, WD, Sinn, DE. 1982. The responses of plankton communities in experimental ponds to atrazine, the most heavily used pesticide in the United States. Ecology 63: 1285-1293.
Detenbeck, NE, Hermanutz, R, Allen, K, Swift, MC. 1996. Fate and effects of the herbicide atrazine in flow-through wetland mesocosms. Environ. Toxicol. Chem. 15: 937-946.
Dewey, Sharon L. 1986. Effects of the herbicide atrazine on aquatic insect community structure and emergence. Ecology 67: 148-162.
Dietz, AC and Schnoor, JL. 2001. Advances in phytoremediation. Environ. Health Perspect. 109: 163-168.
31 Carmen G. Franks - Introduction
Doty, SL, Shang, TQ, Wilson, AM, Tangen, J, Westergreen, AD, Newman, LE, Strand, SE, Gordon, MP. 2000. Enhanced metabolism of halogenated hydrocarbons in transgenic plants containing mammalian cytochrome P450 2E1. Proc. Natl. Acad. Sci. U S A 97: 6287–6291.
Dzantor, EK, Chekol, T, Vough, LR. 2000. Feasibility of using forage grasses and legumes for phytoremediation of organic pollutants. J. Environ. Sci. Health Part A- Toxic/Hazard. Subst. Environ. Eng. 35: 1645-1661.
Fayiga, AO, and Ma, LQ. 2006. Using phosphate rock to immobilize metals in soil and increase arsenic uptake by hyperaccumulator Pteris vittata. Sci. Total Environ. 359: 17-25.
Fent, K, Weston, AA, Caminada, D. 2006. Ecotoxicology of human pharmaceuticals. Aquat. Toxicol. 76: 122-159.
Geng, CN, Zhu, YG, Tong, YP, Christie, P. 2005. Uptake and translocation of arsenic and phosphorous in pho2 mutant and wild type of Arabidopsis thaliana. J. Plant Nut. 28: 1323-1336.
Gfrerer, M., Martens, D, Gawlik, BM, Wenzl, T, Zhang, A, Quan, X, Sun, C, Chen, J, Platzer, B, Lankmayr, E, Kettrup, A. 2002. Triazines in the aquatic systems of the Eastern Chinese Rivers Liao-He and Yangtse. Chemosphere 47: 455-466.
Gorshkov, S, Gorshkova, G, Colorni, B, Gordin, H. 2004. Effects of natural estradiol-17 beta and synthetic 17 alpha-ethynylestradiol on direct feminization of European sea bass Dicentrarchus labrax. J. World Aquacult. Soc. 35: 167-177.
Hashimoto, S, Horiuchi, A, Yoshimoto, T, Nakao, M, Omura, H, Kato, Y, Tanaka, H, Kannan, K, Giesy, JP. 2005. Horizontal and vertical distribution of estrogenic activities in sediments and waters from Tokyo Bay, Japan. Arch. Environ. Contam. Toxicol. 48: 209-216.
Hayes, TB, Collins, A, Lee, M, Mendoza, M, Noriega, N, Stuart, AA, Vonk, A. 2002. Hermaphroditic, demasculinized frogs after exposure to herbicide atrazine at low ecologically relevant doses. Proc. Natl. Acad. Sci. U. S. A. 99: 5476-5480.
Henderson, KLD, Belden, JB, Zhao, SH, Coats, JR. 2006. Phytoremedation of pesticide wastes in soil. Z. Naturforsch. (C) 61: 213-221.
Kartesz, JT. 1994. A Synonymized Checklist of the Vascular Flora of the United States, Canada, and Greenland, Vol. 1, 2nd Edition. Biota of North America Program of the North Carolina Botanical Garden. Timber Press, Portland, OR, USA. 622 p..
Kassel, AG, Ghoshal, D, Goyal, A. 2002. Phytoremediation of trichloroethylene using hybrid poplar. Physiol. Mol. Biol. Plants 8: 3-10.
32 Carmen G. Franks - Introduction
Kavvadias, D, Abou-Mandour, AA, Czygan, FC, Beckmann, H, Sand, P, Riederer, P, Schreier, P. 2000. Identification of benzodiazepines in Artemisia dracunculus and Solanum tuberosum rationalizing their endogenous formation in plant tissue. Biochem. Biophys. Res. Commun. 269: 290-295.
Kleier, DA. 1988. Phloem mobility of xenobiotics. Plant Physiol. 86: 803-810.
Kolpin, DW, Furlong, ET, Meyer, MT, Thurman, EM, Zaugg, SD, Barber, LB, Buxton, HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci. Technol. 36: 1202-1211.
Komori, K, Tanaka, H, Okayasu, Y, Yasojima, M, Sato, C. 2004. Analysis and occurrence of estrogen in wastewater in Japan. Water Sci. Technol. 50: 93-100.
Kuch, HM and Ballschmiter, K. 2001. Determination of endocrine-disrupting phenolic compounds and estrogens in surface and drinking water by HRGC-(NCI)-MS in the picogram per liter range. Envion. Sci. Technol. 35: 3201-3206.
Kuzovkina, YA, Knee, M, Quigley, MF. 2004. Cadmium and copper uptake and translocation in five willow (Salix l.) species. Int. J. Phytorem. 6: 269-287.
Langebartels, C and Harms, H. 1985. Analysis for nonextractable (bound) residues of pentachlorophenol in plant cells using a cell wall fractionation procedure. Ecotoxicol. Environ. Saf. 10: 268-279.
Larsen, M, Ucisik, K, Trapp, S. 2005. Uptake, metabolism, accumulation and toxicity of cyanide in willow trees. Environ. Sci. Technol. 39: 2135-2142.
Lim, PE, Tay, MG, Mak, KY, Mohamed, N. 2003. The effect of heavy metals on nitrogen and oxygen demand removal in constructed wetlands. Sci. Total. Environ. 301: 13-21.
Loffler, D, Rombke, J, Meller, M, Ternes, T. 2005. Environmental fate of pharmaceuticals in water/sediment systems. Environ. Sci. Technol. 39: 5209-5218.
Mathew, R, Kacew, S, Khan, SU. 1998. Bioavailability in rats of bound pesticide residues from tolerant or susceptible varieties of soybean and canola treated with metribuzin or atrazine. Chemosphere 36: 589-596.
Meagher, RB. 2000. Phytoremediation of toxic elemental and organic pollutants. Curr. Opin. Plant Biol. 3: 153-162.
Nehnevajova, E, Herzig, R, Federer, G, Erismann, KH, Schwitzguebel, JP. 2005. Screening of sunflower cultivars for metal phytoextraction in a contaminated field prior to mutagenesis. Int. J. Phytoremediat. 7: 337-349.
33 Carmen G. Franks - Introduction
Nellessen, JE. 2003. Salix exigua Nutt. USDA Forest Service. Online source: http://www.fs.fed.us/global/iitf/pdf/shrubs/Salix%20exigua.pdf
Newman, LA, Strand, SE, Choe, N, Duffy, J, Ekuan, G, Ruszaj, M, Shurtleff, BB, Wilmonth, J, Heilman, P, Gordon, MP. 1997. Uptake and biotransformation of trichloroethylene by hybrid poplars. Envrion. Sci. Technol. 31: 1062-1067.
Nocito, FF, Lancilli, C, Crema, B, Fourcroy, P, Davidian, JC, Sacchi, GA. 2006. Heavy metal stress and sulfate uptake in maize roots. Plant Physiol. 141: 1138-1148.
Orchard, BJ, Doucette, WJ, Chard, JK, Bugbee, B. 2000. Uptake of trichloroethylene by hybrid poplar trees grown hydroponically in flow-through plant growth chambers. Environ. Toxicol. Chem. 19: 895-903.
Pendergrass, A and Butcher, DJ. 2006. Uptake of lead and arsenic in food plants grown in contaminated soil from Barber Orchard, NC. Microchem. J. 83: 14-16.
Purdom, CE, Hardiman, PA, Bye, VJ, Eno, NC, Tyler, CR, Sumpter, JP. 1994. Estrogenic effects of effluent from sewage treatment works. Chem. Ecol. 8: 275- 285.
Quartacci, MF, Argilla, A, Baker, AJM, Navari-Izzo, F. 2006. Phytoextraction of metals from a multiply contaminated soil by Indian mustard. Chemosphere 63: 918-925.
Rankouhi, TR, Sanderson, JT, van Holsteign, I, van Kooten, P, Bosveld, ATC, van den Berg, M. 2005. Effects of environmental and natural estrogens on vitellogenin production in hepatocytes of the brown frog (Rana temporaria). Aquat. Toxicol. 71: 97-101.
Rao, VS. 2000. Principles of Weed Science. 2nd ed. New Hampshire, USA: Science Publishers, Inc.
Richardson, BJ, Larn, PKS, Martin, M. 2005. Emerging chemicals of concern: Pharmaceuticals and personal care products (PPCPs) in Asia, with particular reference to Southern China. Mar. Pollut. Bull. 50: 913-920.
Sandermann, H Jr. 2004. Bound and unextractable pesticidal plant residues: chemical characterization and consumer exposure. Pest Manag. Sci.60: 613-623.
Sandermann, H Jr, Arjmand, M, Gennity, I, Winkler, R. 1990. Animal bioavailability of defined xenobiotic lignin metabolites. J. Agric. Food Chem. 38: 1877-1880.
Sandermann, H Jr, Musick, TJ, Aschbacher, PW. 1992. Animal bioavailability of a 3,4- dichloroaniline-lignin metabolite fraction from wheat. J. Agric. Food Chem. 40: 2001-2007.
34 Carmen G. Franks - Introduction
Sandermann, H Jr. 1994. Higher plant metabolism of xenobiotics: the ‘green liver’ concept. Pharmacogenetics 4: 225-241.
Sapp, M, Ertunc, T, Bringmann, I, Schaffer, A, Schmidt, B. 2004. Characterization of the bound residues of the fungicide cyprodinil formed in plant cell suspension cultures of wheat. Pest Manage. Sci. 60: 65-74.
Sarmah, AK, Northcott, GL, Leusch, FDL, Tremblay, LA. 2006. A survey of endocrine disrupting chemicals (EDCs) in municipal sewage and animal waste effluents in the Waikato region of New Zealand. Sci. Total Environ. 355: 135-144.
Scheunert, I, Topp, E, Schmitzer, J, Klein, W, Korte, F. 1985. Formation and fate of bound residues of [14C]benzene and [14C]chlorobenzenes in soil and plants. Ecotoxicol. Environ. Saf. 9: 159-170.
Schnoor, J. 2002. GWRTAC Technology Evaluation Report. Phytoremediation of Soil and Groundwater. Iowa City: The University of Iowa. Report nr E-Series: TE-02- 01. 45 p. Available online: www.gwrtac.org
Schultz, IR, Skillman, A, Nicolas, JM, Cyr, DG, Nagler, JJ. 2003. Short-term exposure to 17 alpha-ethynylestradiol decreases the fertility of sexually maturing male rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 22: 1272-1280.
Sengl, M and Krezmer, S. 2003. Proficiency tests for pharmaceuticals in different waters. Accred. Qual. Assur. 8: 523-529.
Servos, MR, Bennie, DT, Burnison, BK, Jurkovic, A, McInnis, R, Neheli, T, Schnell, A, Seto, P, Smyth, SA, Ternes, TA. 2005. Distribution of estrogens, 17β-estradiol and estrone, in Canadian municipal wastewater treatment plants. Sci. Total Environ. 336: 155-170.
Shen, H, Christie, P, Li, X. 2006. Uptake of zinc, cadmium and phosphorus by arbuscular mycorrhizal maize (Zea mays L.) from a low available phosphorus calcareous soil spiked with zinc and cadmium. Environ. Geochem. Health 28: 111-119.
Shone, MGT and Wood, AV. 1974. A comparison of the uptake and translocation of some organic herbicides and a systemic fungicide by barley. I. Adsorption in relation to physico-chemical properties. J. Exp. Bot. 25: 390-400.
Skidmore, MW, Paulson, GC, Kuiper, HA, Ohlin, B, Reynolds, S. 1998. Bound xenobiotic residues in food commodities of plant and animal origin. IUPAC Reports on Pesticides (40). Pure Appl. Chem. 70: 1423-1447. Sosiak, A and Hebben, T. 2005. A preliminary survey of pharmaceuticals and endocrine disrupting compounds in treated municipal wastewaters and receiving waters of Alberta. Environmental Monitoring and Evaluation Branch. Alberta Environment.
35 Carmen G. Franks - Introduction
Pub. No: T/773. 64 p. Available online: http://www3.gov.ab.ca/env/info/infocentre/publist.cfm
Stan, HJ and Linkerhägner, M. 1992. Vom Wasser 79: 75-88.
Sterky, F, Bhalerao, RR, Unneberg, P, Segerman, B, Nilsson, P, Brunner, AM, Charbonnel-Campaa, L, Lindvall, JJ, Tandre, K, Strauss, SH, Sundberg, B, Gustafsson, P, Uhlen, M, Bhalerao, RP, Nilsson, O, Sandberg, G, Karlsson, J, Lundeberg, J, Jansson, S. 2004. A Populus EST resource for plant functional genomics. Proc. Natl. Acad. Sci. U. S. A. 101: 13951-13956.
Stevens, M, Fenchel, G, Hoag, C, Anderson, MK. 2000. NRCS Plant Guide Coyote Willow.
Subramanian, M, Oliver, DJ, Shanks, JV. 2006. TNT phytotransformation pathway characteristics in Arabidopsis: role of aromatic hydroxylamines. Biotechnol. Prog. 22: 208-216.
Taiz, L and Zeiger, E. 2002. Plant physiology. 3rd ed. Sunderland, MA: Sinauer Associates, Inc.
Tang, JX, Hoagland, KD, Siegfried, BD. 1997. Differential toxicity of atrazine to selected freshwater algae. Bull. Environ. Contam. Toxicol. 59: 631-637.
Ternes, TA. 1998. Occurrence of drugs in German sewage treatment plants and rivers. Wat. Res. 32: 3245-3260).
Ternes, TA, Kreckel, P, Mueller, J. 1999b. Behaviour and occurrence of estrogens in municipal sewage treatment plans – II. Aerobic batch experiments with activated sludge. Sci. Total Environ. 225: 91-99.
Ternes, TA, Stumpf, M, Mueller, J, Haberer, K, Wilken, R-D, Servos, M. 1999a. Behavior and occurrence of estrogens in municipal sewage treatment plants – I. Investigations in Germany, Canada and Brazil. Sci. Total Environ. 225: 81-90.
The Arabidopsis Genome Initiative. 2000. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796-815.
Thompson, PL, Ramer, LA, Schnoor, JL. 1998. Uptake and transformation of TNT by hybrid poplar trees. Environ. Sci. Technol. 32: 975-980.
Tracy, TS. 2004. Drug absorption and distribution. In: Craig, CR and Stitzel, RE. Modern pharmacology with clinical applications. 6th ed. Philadelphia: Lippincott Williams & Wilkins. p 20-33.
36 Carmen G. Franks - Introduction
Trapp, S. 2004. Plant uptake and transport models for neutral and ionic chemicals. Environ. Sci. Pollut. Res. 11: 33-39.
Trapp, S, Ciucani, G, Sismilich, M. 2004. Toxicity of tributyltin to willow trees. Environ. Sci. Pollut. Res. 11: 327-330.
Tyerman, SD, Niemietz, CM, Bramley, H. 2002. Plant aquaporins: multifunctional water and solute channels with expanding roles. Plant Cell Environ. 25: 173-194.
Unseld, E, Kirshna, DR, Fischer, C, Klotz, U. 1989. Detection of desmethyldiazepam and diazepam in brain of different species and plants. Biochem. Pharmacol. 38: 2473- 2478. van der Ven, K, Van Dongen, W, Maes, BUW, Esmans, EL, Blust, R, De Coen, WM. 2004. Determination of diazepam in aquatic samples by capillary liquid chromatography-electrospray tandem mass spectrometry. Chemosphere 57: 967- 973.
Wayne, R and Tazawa, M. 1990. Nature of water channels in the internodal cells of Nitellopsis. J. Membr. Biol. 116: 31-39.
Weiss, JM, McKay, AJ, DeRito, C, Watanabe, C, Thorn, KA, Madsen, EL. 2004. Development and application of pyrolysis gas chromatography/mass spectrometry for the analysis of bound trinitrotoluene residues in soil. Environ. Sci. Technol. 38: 2167-2174.
Werner, J, Wautier, K, Evans, RE, Baron, CL, Kidd, K, Palace, V. 2003. Waterborne ethynylestradiol induces vitellogenin and alters metallothionein expression in lake trout (Salvelinus namaycush). Aquat. Toxicol. 62: 321-328.
Wildmann, J. 1988. Increase of natural benzodiazepines in wheat and potato during germination. Biochem. Biophys. Res. Commun. 157: 1436-1443.
Yang, X, Feng, Y, He, Z, Stoffella, PJ. 2005. Molecular mechanisms of heavy metal hyperaccumulation and phytoremediation. J. Trace Elem. Med. Biol. 18: 339-353.
Zebrowski, W, Buszewski, B, Lankmayr, E. 2004. Modeling of uptake of xenobiotics in plants. Crit. Rev. Anal. Chem. 34: 147-164.
Zuccato, E, Chiabrando, C, Castiglioni, S, Calamari, D, Bagnati, R, Schiarea, S, Fanelli, R. 2005. Cocaine in surface waters: a new evidence-based tool to monitor community drug abuse. Environmental Health: A Global Access Science Source 4. p 7.
37 Carmen G. Franks - Salix exigua Phytoremediation
CHAPTER 2 Phytoremediation of trace levels of pharmaceuticals, diltiazem,
diazepam and 17α-ethynylestradiol, and the herbicide atrazine, with sandbar willow
(Salix exigua)
2.1 INTRODUCTION
The discovery that many biologically active compounds leave our wastewater treatment
plants and enter the environment (Ternes et al., 1999; Kolpin, et al., 2002; Sosiak and
Hebben, 2005), and occur in our drinking water (Stan and Linkerhägner, 1992, cited in
Sengl and Krezmer, 2003; Kuch and Ballschmiter, 2001), has led to environmental and
human health concerns. The environmental fate of these compounds, particularly
pharmaceuticals, is poorly understood.
Phytoremediation, the use of plants to remediate environmental contaminants, is not new,
but the term was only introduced in the scientific literature in about 1994 (Schnoor,
2002). Currently used for the removal of hydrocarbons, heavy metals and pesticides,
phytoremediation may provide a natural and cost-effective method of removing wastewater pharmaceuticals from our waterways and thus minimizing their ecological and health impacts. The prospective phytoremediation of pharmaceuticals represents a new direction for research and application that should be considered, particularly as the quantities of contaminants are only increasing with increasing and aging populations and the proliferation of drug-treatments for human health.
38 Carmen G. Franks - Salix exigua Phytoremediation
Willow species, members of family Salicaceae and genus Salix, are rarely reported for
phytoremediation projects, but they satisfy many requirements of a plant for
phytoremediation. They are easy to propagate and grow rapidly and have abundant water
usage with roots extending into water tables and may thus be especially suitable for
phytoremediation of contaminants in ground and surface waters, including streams that
often provide the primary water sources for municipal use. Willow also has the ability to
survive prolonged periods of inundation, a characteristic useful when the areas to be
remediated are riparian zones along river banks or lake shores. Salix exigua, commonly
referred to as sandbar willow, was chosen for this project due to its local availability,
extensive distribution and abundance, and preference for moist zones immediately
adjacent to streams.
The present investigation studied the potential of this prominent riparian shrub to uptake
and transport three common wastewater pharmaceuticals: diltiazem (DTZ), a calcium
channel blocker; diazepam (Valium®) (DZP), an antianxiety drug; and 17α-
ethynylestradiol (EE2), a synthetic birth control hormone that is a common component of
the contraceptive pill. The herbicide atrazine (ATZ) was used as a positive control for
this project, since it is known to be readily taken up by plants and is another prominent
water pollutant that has been extensively investigated relative to phytoremediation.
Uptake and transport studies were undertaken, and subsequently root concentration factor
(RCF) analyses were performed, to enable a comparison of Salix exigua root and
39 Carmen G. Franks - Salix exigua Phytoremediation pharmaceutical relationships with other documented compound and plant root relationships, particularly for uptake studies with barley and hybrid poplar.
40 Carmen G. Franks - Salix exigua Phytoremediation
2.2 MATERIALS AND METHODS
2.2.1 Chemicals
The compounds chosen for this study included the synthetic hormone 17α-
ethynylestradiol (an EDC) (EE2), the anti-hypertensive agent diltiazem (DTZ), the
anticonvulsant diazepam (DZP) and the herbicide atrazine (ATZ) (Figure 2.1). Cis-(+)-
[N-methyl-3H]-diltiazem (specific activity: 74.5 Ci mmol-1; radiochemical purity > 97%
to HPLC analysis), [methyl-3H]-diazepam (specific activity: 86.0 Ci mmol-1; radiochemical purity > 97% to HPLC analysis), and 17α-[6,7-3H(N)]-ethynylestradiol
(specific activity: 40.0 Ci mmol-1; radiochemical purity >97% to HPLC analysis) were
obtained from PerkinElmer Life Sciences, Inc. (Boston, MA, USA). [Ring-U-14C]-
atrazine (specific activity: 10.35 mCi mmol-1; radiochemical purity > 95% to HPLC
analysis) was obtained from Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada). Non-
labeled DTZ-hydrochloride, DZP, ATZ, and EE2 were obtained from Sigma-Aldrich
Canada Ltd. Standards were dissolved in ethanol, since methanol is toxic to plants, and
to maintain consistent experimental conditions since EE2 and DZP are insoluble in water
but DTZ and ATZ are water-soluble. Some chemical properties of the compounds are
listed in Table 2.1.
41 Carmen G. Franks - Salix exigua Phytoremediation
Diltiazem HCl + OCH Atra zine 3
Cl H N N S CH H - H 3 Cl
NHC OOCCH3 CHHN25 N CH N 3 O 2-c hloro-4(e thyla m ino)-6-(isop rop yla m ino )-s-tria zine CH22 CH N(CH 32 )
1,5-Benzothiazepin-4(5H)-one,3-(acetyloxy)-5[2-(dimethylamino)ethyl] -2,-3- dihydro-2(4-methoxyphenyl)-, mono-hydrochloride, (+ )-cis
Diazepam (Valium® ) HC 3 O 17α -Ethynylestradiol OH N CH3 C CH
H Cl N H H
HO 17-ethynyl-13-methyl-7,8,9,11,12,13,14,15,16,17 7-chloro-1-methyl-5-phenyl-2H-1,4-benzodiazepin-2-one -decahydro-6H-cyclopenta[a]phenanthrene-3,17-diol
Figure 2.1. Chemical structures of the three pharmaceuticals and the herbicide used in the present phytoremediation study.
42 Carmen G. Franks - Salix exigua Phytoremediation
Table 2.1. Chemical properties of the three pharmaceuticals and the herbicide used in the present phytoremediation study.
M.W. CAS Number Log K * Water Solubility (g/ mol) ow
17α-ethynylestradiol 57-63-6 296.4 3.67 Slightly Soluble
diazepam 439-14-5 284.8 2.82 Slightly Soluble
diltiazem 42399-41-7 451.0 2.70 Soluble
atrazine 1912-24-9 215.7 2.61 Slightly Insoluble
TM * Log Kow values estimated from ECOSAR (online demo source: http://www.syrres.com/esc/est_kowdemo.htm).
43 Carmen G. Franks - Salix exigua Phytoremediation
2.2.2 Plants, hydroponics and treatments
Sandbar willow (Salix exigua Nutt.) stem cuttings were collected from riparian zones in
late winter and early spring along the Oldman River, Lethbridge, Alberta. The cuttings
were about 0.5 cm in diameter and 10 cm long, and placed in water to allow adventitious
rooting. Once rooting had begun the cuttings were transferred to a hydroponic system in a growth chamber.
The hydroponic system was adapted from Gibeaut et al. (1997) and used 37.9 L opaque
tubs measuring 60.1 L x 47.0 W x 22.2 H cm for the reservoir. Forty 1.5 cm holes were
drilled in the opaque plastic lid. A continuous aeration system used an aquarium pump
(Petcetera Air Pump AP-3800) and 12.7 cm long diffusing stone. Reverse osmosis
purified water was used enabling the pH of the hydroponic solution to drop to slightly
acidic upon addition of Dutch Nutrient Formula Gro A & B solution, at 2.5 mL/ L. The
rooted cuttings were inserted through the holes drilled into the reservoir lid and a strip of
horticultural Rock Wool made from basalt fibers (FibrGrow Horticultural Products,
Sarnia, Ontario), was wrapped around the cutting to provide support and block light from
the reservoir (Figure 2.2).
Growth chamber (Conviron Model E15, Controlled Environments Ltd., Winnipeg, MB,
Canada) conditions for willow were set at 16 hour day/ 8 hour night photoperiod,
temperature was maintained at 20 °C, relative humidity was 70%, and photosynthetically
active radiation at plant height was approximately 306 μmol s-1m-2 provided by a mixture
of halogen 400W bulbs (Sylvania Metalarc and Philips). Willow cuttings approximately
44 Carmen G. Franks - Salix exigua Phytoremediation
20 days old with an approximate shoot length range of 8 to 12 cm were used for the uptake studies.
45 Carmen G. Franks - Salix exigua Phytoremediation
Figure 2.2. Hydroponic system with willow cuttings.
46 Carmen G. Franks - Salix exigua Phytoremediation
2.2.3 Uptake studies
Uptake studies were performed for 24 hours with solutions in 30 mL glass culture tubes
using the hydroponically grown willow cuttings. This duration was sufficient for
considerable uptake for analysis and to allow the measurement of rate of uptake. The 30 mL culture tubes were sufficient in size for the root mass of the willow cuttings and transpiration.
For each chemical replicate, 1 μg of the unlabeled compound plus 16,667 Bq of the
radiolabeled compound was added to each culture tube (in ethanol solution resulting in
4.2 μL/ mL). Nutrient solution (24 mL) was taken from the active hydroponic reservoirs
and added to the culture tubes for a concentration of 0.04 μg/ mL or 40 ng/ mL. At this
time (t0) a sample was taken and analyzed by a liquid scintillation counting (LSC).
Similar sized plants were chosen and inserted into the culture tubes. A rooted cutting was
inserted into the culture tube at a level maintaining the stem cutting out of solution, with
plant shoots supported, so that most of the root mass was submerged (Figure 2.3). The
tubes were wrapped in foil and the study was run with the hydroponic growth chamber
settings described above. Samples taken directly from the culture tube solution were
analyzed with LSC at 2, 4, 8, and 24 hours to determine uptake, represented as loss of
radioactivity from the solution. At harvest (24 hr) the cuttings were removed from
solution, the roots were rinsed in nutrient solution to remove surface solution on the
roots, and divided into green shoot (leaves and green stem), stem cutting, and roots. Each
was weighed fresh and stored at -20 °C until analysis. Replicate plant numbers were 6 for
47 Carmen G. Franks - Salix exigua Phytoremediation
EE2, DZP, and ATZ, and 4 for DTZ. The experiment was repeated twice for EE2, DZP and ATZ, but not for DTZ as the DTZ standard was severely degraded by the time of the
second experiment.
Transpiration volumes were monitored at each sampling time to check for changes in
transpiration rates (a possible sign of phytotoxicity) and to provide a comparison between
plants and compounds, as well as for estimating the transpiration stream concentration
factors (TSCF). Just prior to sampling, nutrient solution was added to the culture tube to replace water lost through transpiration. The volume added was recorded as the volume transpired over the sampling period. The solution was mixed and allowed to sit for several minutes to ensure mixing of original and new solutions.
Other. Blank runs were monitored for the 24 hour period to determine evaporation,
volatility and binding of the compounds to the glassware. For this control, compounds
were added to nutrient solution, the tubes were covered with foil and the solution
sampled at the same time points to measure remaining radioactivity.
Testing for compound binding to the culture tubes involved drying down the remaining
solution after the termination of the study and rinsing the tubes with methanol. The
concentrated solution plus methanol rinse was then resuspended and analyzed by LSC.
48 Carmen G. Franks - Salix exigua Phytoremediation
Figure 2.3. Culture tube set up for analysis of pharmaceuticals by willow.
49 Carmen G. Franks - Salix exigua Phytoremediation
2.2.4 Root concentration factors and transpiration stream concentration factors
To determine root concentration factors (RCF) for Salix exigua, 24 hour studies were
performed using excised roots and roots attached to the cutting but without a shoot. RCF studies were carried out in nutrient solution in 30 mL glass culture tubes using hydroponically grown willow cuttings. For each chemical replicate, 1 μg (0.04 μg/ mL) of the unlabeled compound plus 8333 Bq of the radiolabeled compound was added to each culture tube (in ethanol solution). RCF values were only determined for EE2, DZP and ATZ, as the radioactive DTZ stock was degraded at the time of this study. Nutrient solution (24 mL) was taken from the active hydroponic reservoirs and added to the culture tubes. At this time (t0) a sample was taken and analyzed by LSC.
For these analyses, willow plants were chosen based on apparently similar root masses.
The roots were severed from the plant and submerged in the test tube solutions. For the
roots that were left attached to the cutting the green shoot was removed and the roots
submerged in the culture tube solution while maintaining the cutting out of the solution, as with the uptake studies. The test tubes were covered with foil to block out light and these studies were performed under the same environmental conditions as the uptake studies described in section 2.2.3. The experiment was performed with three replicates of excised roots and three replicates of roots attached to the cutting, for each of EE2, DZP and ATZ.
As with the uptake studies, the solution was sampled and analyzed by LSC to determine
uptake, represented as removal of radioactivity from solution. The solutions were
50 Carmen G. Franks - Salix exigua Phytoremediation sampled at 1, 2, 4, 8 and 24 hours and each sampling was followed by the addition of nutrient solution to replace the volume removed due to sampling. Evaporative losses were negligible.
At 24 hours the study was terminated. The excised roots were rinsed in fresh nutrient solution from the active hydroponic systems, blotted dry and weighed fresh. The same was carried out for roots attached to the cutting, with roots and wood being separated for individual weighing. Roots and cutting were stored at -20°C until solvent extraction and oxidation could be carried out. RCF values were calculated using results from excised roots and roots attached to the cutting, and estimated using results from whole plant uptake studies, and the equation (Shone and Wood, 1974):
root uptake = RCF (μg/mL) x root mass (g) x external solution concentration
(μg/mL).
RCF for the compounds were also calculated from log Kow values using equations developed by Briggs et al. (1982) for barley roots:
log (RCF - 0.82) = 0.77 log Kow - 1.52,
and by Burken and Schnoor (1998), for hybrid poplar roots:
log (RCF - 3.0) = 0.65 log Kow – 1.57.
51 Carmen G. Franks - Salix exigua Phytoremediation
Transpiration stream concentration factor (TSCF) for the four compounds in willow were estimated using results from whole plant uptake studies and the equation (Briggs et al.,
1982):
TSCF = [concentration in the shoot (μg) / volume transpired (mL)] / [(external solution concentration (μg/mL)initial + external solution concentration (μg/mL)final )/
2].
Although transpiration may not have been continuous, the means of initial and final external solution concentrations were used to estimate overall TSCF (Briggs et al., 1982).
TSCF for the compounds were also calculated from log Kow values using equations
developed by Briggs et al. (1982) for barley shoots:
2 log TSCF = 0.784 exp[-(log Kow – 1.78) / 2.44],
and by Burken and Schnoor (1998), for hybrid poplar shoots:
2 log TSCF = 0.756 exp[-(log Kow -2.50) / 2.58].
52 Carmen G. Franks - Salix exigua Phytoremediation
2.2.5 Soluble fractions
For half of roots and shoots harvested at the end of the uptake studies, except for DTZ in
which all roots were used, a methanol solvent extraction was used to separate the soluble
fraction from the non-extractable fraction that may have been contained within the cell
walls. The determination of soluble versus bound fractions provides insight into the fate
of the compounds once they are in the plant. Soluble fractions may still have the potential
to be transported or metabolized, while bound fractions may be immobile, with minimal
degradation after binding.
For roots, the frozen sample was ground in a mortar to a fine pulp with 80% aqueous
methanol (MeOH:H2O 80:20), centrifuged, and the supernatant decanted. The residue
was resuspended with 100% methanol, centrifuged, decanted and the extraction was
repeated a third time. The supernatants were pooled, aliquots taken and bleached using
commercially available sodium hypochlorite (5.25% chlorine), and analyzed by LSC.
The solvent extraction procedure was repeated for the green shoot following a similar
extraction and bleaching procedure. Depending on the mass of the shoot and leaves an extra volume of methanol was added and the steps were repeated a fourth time until the pigmentation was removed from the residue.
Initially, the cutting of the willows was ground to a powder in a mortar with sand and
liquid nitrogen. Additional grinding was performed using 80% aqueous methanol. The
sample was then cut and homogenized with sonication using a Polytron (Kinematica CH-
53 Carmen G. Franks - Salix exigua Phytoremediation
6010, Brinkmann Instruments, Westbury, NY). The homogenized sample was then
subjected to a similar extraction procedure as the shoots. The extract was sampled and
bleached for analysis by LSC, but it was found to contain concentrations of lignin that
could not be bleached and which interfered with LSC. Subsequently, the cuttings were
oxidized to determine total radioactivity that included both bound and soluble fractions.
Initially, the purification procedure involved a solid phase extraction (SPE) step using octadecyl-functionalized silica (C18) to remove pigments from coloured samples. The
supernatants were pooled and adjusted to 80% aqueous methanol and pH 7 using dilute
sodium hydroxide. The extract was then run through a 2 g - C18 column to remove a large
percentage of the pigments that interfere with LSC. Aqueous methanol (80%) was used to
elute the compounds of interest, typically requiring less than 10 mL to remove it from the
column, but this volume varied with compounds. The eluant was then sampled and
analyzed using LSC.
Results from this purification procedure suggested there might be losses of radioactivity
within either the C18 or from binding to particulate matter, notably with DTZ and EE2.
The procedure was modified, removing the C18 clean-up step. Instead, the pooled
supernatant was collected green and aliquots taken to be bleached prior to LSC as
described above. Addition of bleach to the sample was sufficient to remove the color
while not affecting the counting efficiency. Appendix B provides some of the results from the purification procedure that suggest the loss of radioactivity with the SPE procedure.
54 Carmen G. Franks - Salix exigua Phytoremediation
2.2.6 Bound fractions
Some of the compounds taken into the plants undergo a process that bind them
irreversibly to cell wall constituents and these are then typically termed as ‘bound.’ This
bound fraction is characterized by the inability to extract it using an extensive solvent
extraction. Such bound radioactivity that could not be removed with organic solvent was
quantified through oxidation that was performed on willow residues remaining after
solvent extraction. As well, several replicate plants were analyzed whole without the
prior solvent extraction to confirm the recovery levels from combination of the soluble
extract fractions and the bound residue oxidation fractions.
Whole plant fractions and plant residues containing bound radioactivity were combusted
in an R.J. Harvey Instrument Corporation Biological Material Oxidizer OX-500.
14 Trapping of CO2 from the oxidizer was done using the CO2 absorber Carbo-Sorb E and the LSC cocktail Permafluor E+ obtained from PerkinElmer Life and Analytical
3 Sciences, Inc. (Boston, MA, USA). Trapping of H2O from the oxidizer was done using
the scintillation cocktail Ecolite manufactured by MP (formerly ICN). CarboSorb E was
combined with Permafluor E+ in a ratio of 1:2 in a trapping volume of 15 mL. Fractions
of the whole plant of sufficient size (0.50 – 0.75 g fresh weight for roots and shoots, 0.20
g fresh weight for cutting) were oxidized to determine entire sample radioactivity that
was analyzed by LSC. Plant residues were weighed dry (< 0.25 g dry weight) and
oxidized after the addition of 0.2 to 0.5 mL of water.
55 Carmen G. Franks - Salix exigua Phytoremediation
Prior to oxidation, two procedures, cell wall fractionation and cell wall dissolution, were attempted to extract the bound residue for LSC. Both procedures are outlined in
Appendix B. Neither procedure was able to provide results that could be reliably analyzed by LSC due to extensive pigmentation.
56 Carmen G. Franks - Salix exigua Phytoremediation
2.3 RESULTS
2.3.1 Uptake studies time course
Uptake study experiments were replicated twice, excluding DTZ due to degraded
samples. Similar results were obtained from both replicate experiments and only one data
set is presented for DZP, ATZ and EE2.
Percentage uptake, or removal from solution over time for EE2, DZP and DTZ followed a very similar pattern although maximum values varied between 10 and 30% (Figure
2.4). ATZ uptake followed a flatter curve than the other compounds, but uptake at 24 hours was 6% lower than for DZP (Figure 2.4). Uptake within the first 2 hours was generally consistent among the three pharmaceuticals, reaching approximately 37 - 50% removal from solution. ATZ was removed more slowly than the pharmaceuticals with about 13.5% uptake within the first 2 hours (Figure 2.4).
Salix exigua was thus very effective in removing EE2, DTZ and DZP from solution with
removal of 88, 77, and 56%, respectively, in 24 hours. About 50% of the herbicide ATZ
was removed within the 24 hours with the same treatment conditions.
Among compounds, the plants’ weights (root, cutting, shoot and total weight) and total
volume transpired, did not vary significantly (Table 2.2) (Table A2.1, ANOVA, p > 0.05) except for the cutting (ANOVA, p = 0.004); thus, replicate plants were of similar size and apparently similar health.
57 Carmen G. Franks - Salix exigua Phytoremediation
No plants showed signs of phytotoxicity during the period of study and with the
concentrations used. There was no observable discoloration of the leaves or significant
alteration in transpiration rates that were approximated by straight lines of cumulative
transpiration up to 8 hours (Figure 2.5). A slight reduction in transpiration rate between 8
and 24 hours probably reflects reduced transpiration during the night phase of the study.
Similar cumulative transpiration trends over the 24 hour period were observed for all
replicate plants for all compounds (Figure 2.5), but cumulative volumes transpired at t =
2, 4, and 8 did vary significantly between some of the compounds (Table A2.2, ANOVA,
p < 0.05). At t = 2, volumes transpired varied significantly between DTZ or ATZ and
EE2, but not between ATZ and EE2 (Table A2.3, Dunnett’s C, p < 0.05). At t = 4 and t =
8, volumes transpired varied significantly between ATZ and DTZ (Table A2.3, Dunnett’s
C, p < 0.05).
A correlation analysis (Table A2.4, Spearman’s rho) for uptake with fresh weight and
volume transpired at each sampling time, found no consistent significant correlation
between these factors and uptake for the compounds. ATZ uptake was only correlated
with root and shoot fresh weight, and cumulative transpiration at t = 24 (Spearman’s rho,
p < 0.05). DZP uptake was correlated with shoot or total fresh weight, and cumulative
volume transpired at t = 8 and t = 24 (Spearman’s rho, p < 0.05). DTZ uptake appeared to be correlated to shoot, root and total fresh weight, and cumulative volume transpired at t
= 8 and t = 24 (Figure A2.1). EE2 uptake was significantly correlated with root fresh
weight at t = 4 and t = 24 (Spearman’s rho, p < 0.05).
58 Carmen G. Franks - Salix exigua Phytoremediation
Cumulative volume transpired was consistently correlated with shoot and total fresh weights for DZP and EE2 over the study period (Table A2.5, Spearman’s rho, p < 0.05).
Cumulative volume transpired for ATZ was consistently correlated with total fresh weight across the sample times (p < 0.05).
Volatility of the compounds under the growth chamber conditions was negligible for the four compounds in the plant-free control conditions. Similarly, testing for compound binding to the culture tubes revealed no remaining glassware-bound radioactivity up to the t = 24 sampling time. It would be assumed that binding to the glassware would have occurred rapidly and therefore would not account for the observed uptake curves.
59 Carmen G. Franks - Salix exigua Phytoremediation
100
75
50
25
17α-Ethynylestradiol 0 100
75
50
25
Diltiazem 0 100 Uptake (%)
75
50
25
Diazepam
0 100
75
50
25
Atrazine
0 0 4 8 12 16 20 24 Time (hours)
Figure 2.4. Percentage uptake from solution of 0.04 μg/ mL for 17α-ethynylestradiol (n
= 6), diltiazem (n = 4), diazepam (n = 6), and atrazine (n = 6), by Salix exigua. Mean ±
SE are plotted for the 24 hour period. The 16 hr day and 8 hr night (shaded) are indicated.
60 Carmen G. Franks - Salix exigua Phytoremediation
Table 2.2. Salix exigua summary plant information from uptake study. Mean ± SE are shown for plant fresh weights and total volume transpired.
Root Cutting Shoot Total Volume n fr. wt. (g) fr. wt. (g) fr. wt. (g) Transpired (mL)
17α-Ethynylestradiol 6 0.47 ± 0.05 0.91 ± 0.05 0.71 ± 0.05 10.42 ± 0.58
Diltiazem 4 0.41 ± 0.13 0.61 ± 0.05 0.64 ± 0.08 9.25 ± 0.46
Diazepam 6 0.53 ± 0.10 0.76 ± 0.09 0.93 ± 0.16 13.67 ± 2.39
Atrazine 6 0.48 ± 0.08 1.23 ± 0.15 1.00 ± 0.09 14.25 ± 1.23
61 Carmen G. Franks - Salix exigua Phytoremediation
20
15
10
5
17α-Ethynylestradiol 0 20
15
) 10 mL (
5
Diltiazem 0 ired volume ired volume 20 p
15
10 Cumulative trans
5 Diazepam 0 20
15
10
5
Atrazine 0 04812162024 Time (hours)
Figure 2.5. Cumulative transpired volumes measured for Salix exigua at each sampling time for the four compounds. Mean ± SE are shown. The 16 hr day and 8 hr night
(shaded) are indicated.
62 Carmen G. Franks - Salix exigua Phytoremediation
2.3.2 Distribution
Plant components underwent extensive solvent extraction to retrieve the soluble fraction
and the remaining residue was oxidized to retrieve the bound fraction. Recovery levels of
radioactivity removed from solution were somewhat variable across the replicates and especially across compounds. Final percent recoveries for EE2, DTZ, DZP and ATZ were 88, 94, 80 and 37%, respectively (Figures 2.6 and 2.7).
EE2 recovery was mostly as a bound fraction within the roots with a small soluble
fraction (10%), making up 92% of recovered radioactivity (Figures 2.6 and 2.7). Shoot
soluble and bound fractions made up 6% of recovered with the bound fraction making up
less than 1% of this value. Approximately 2% of EE2 was within the cutting. DTZ was
recovered entirely as a soluble fraction within the roots (Figures 2.6 and 2.7). DZP
remained primarily soluble with proportion of recovered radioactivity at 50% in the
shoot, 14% in wood and 38% within the roots (Figures 2.6 and 2.7). The bound fraction
represented less than 1% of total recovered radioactivity.
ATZ was not significantly bound to roots or shoots of willow. The recovery of
radioactivity after extraction of shoots was low (Figure 2.7), but from oxidation the
largest portion of recovered radioactivity was within the shoots (68%) (Figure 2.6). The
cutting proportion of the recovered radioactivity was 20% and the roots contained 12%.
63 Carmen G. Franks - Salix exigua Phytoremediation
100 Root Wood Shoot Total Plant
50 Recovery (%) Recovery
0 17alpha- Diltiazem Diazepam Atrazine Ethynylestradiol
Figure 2.6. Distribution of recovered radioactivity from oxidation of whole root, shoot and cuttings and total plant recovery for radiolabeled compound added to Salix exigua for
17α-ethynylestradiol, diazepam, and atrazine (n = 3 for each) (mean ± SE). For diltiazem, n = 4, root value is from the soluble fraction recovery as analysis of shoot, wood and
bound fractions recovered no measurable amounts of radioactivity.
64 Carmen G. Franks - Salix exigua Phytoremediation
100 Root Soluble
Root Bound
Cutting 75 Shoot Soluble
Shoot Bound
50 Recovery (%)
25
0 17alpha- Diltiazem Diazepam Atrazine Ethynylestradiol
Figure 2.7. Distribution of recovered radioactivity among soluble and bound fractions in roots and shoots and cutting for radiolabeled compound added to Salix exigua (mean ±
SE). For 17α-ethynylestradiol, diazepam, and atrazine, n = 6; 3 whole plant oxidized, 3 soluble + bound oxidized. For diltiazem, n = 4, soluble + bound oxidized.
65 Carmen G. Franks - Salix exigua Phytoremediation
2.3.3 Root concentration factors and transpiration stream concentration factors
Root concentration factor (RCF) experiments were carried out for Salix exigua to provide
a comparison with other plants and to establish the relationship between RCF and
pharmaceutical log Kow values for willow roots. The experiment was terminated at 24
hours when there was apparently compound equilibrium between roots and solution. In
prior studies, equilibrium apparently occurred quickly, within hours (Burken and
Schnoor, 1998) or up to 24 hours (Briggs et al., 1982).
Fresh weights of the two types of roots used for determination of RCF values were
similar, allowing comparison of uptake results between the two (Table A2.6, ANOVA, p
> 0.05). Root uptake with excised roots followed similar trends for EE2, DZP and ATZ.
Final root uptakes for EE2, DZP and ATZ were 79, 42 and 31%, respectively. Root
uptake with roots attached to the cutting for the 3 compounds also followed similar
patterns of uptake, although final root uptakes were slightly higher, at 88, 54 and 41%,
respectively. The roots attached to the cuttings had uptake curves similar to excised roots,
but uptake was faster for roots attached to the cutting and significantly different with 7 of
15 sampling pairs (Figure 2.8; Table A2.7, ANOVA, p < 0.05).
Average root uptake of results combined for the two root types varied significantly between compounds for all sampling times except t = 1 (Table A2.8, ANOVA, p > 0.05).
At t = 2, EE2 and DZP root uptake varied significantly (Table A2.9, Tamhane, p =
0.000). By t = 4, EE2 and ATZ varied significantly (Tamhane, p = 0.029). At t = 8 and t
66 Carmen G. Franks - Salix exigua Phytoremediation
= 24, all compounds were significantly different in their root uptake except for ATZ and
DZP.
RCF values for excised roots and roots attached to the cutting were calculated separately, but were similar enough that one equation was sufficient to explain the relationship between roots and log Kow (Table A2.7, ANOVA). DTZ was excluded from this equation since the RCF value was calculated using whole plant uptake values. The equation developed from these calculations with an R2 = 1.00 for willow RCF was (Figure 2.9):
log (RCF) = 0.93 log Kow – 1.10.
RCF values were also calculated for the compounds using their log Kow values and the equations for barley (Briggs et al. 1982) and poplar roots (Burken and Schnoor, 1998)
(Figure 2.9).
RCF plants underwent similar solvent extraction and oxidation as for the whole plant uptake studies. The average percent recoveries and distributions between the excised roots and roots attached to the cutting varied among replicates and between compounds
(Figure 2.10). Average percent recovery for EE2, DZP and ATZ were 54, 40 and 10%, respectively, for combined results from excised roots and roots attached to cuttings
(Figure 2.10). The distribution ratios within the roots and wood were generally similar to those observed from the roots of whole plant uptake studies. Thus, soluble fractions contained more than 85% of recovered radioactivity for DZP and ATZ, while recovered
67 Carmen G. Franks - Salix exigua Phytoremediation radioactivity for EE2 was more than 70% in the bound fraction (Figure 2.10). For all compounds, proportion within the cutting was small.
The RCF values for the four compounds and willow were estimated using results from the whole plant uptake studies and the equation for RCF (Shone and Wood, 1974).
Calculated RCF values were not significantly different from experimentally derived RCF values except for ATZ (t-test, t = -3.25, df = 5, p = 0.023).
68 Carmen G. Franks - Salix exigua Phytoremediation
100 * *
75 *
50
25 Roots+Wood 17α-Ethynylestradiol Roots 0 100 (%) Diazepam Roots+Wood Roots lution
o 75
* l from s 50 a * *
Remov 25
0 100 Atrazine Roots+Wood Roots 75 *
50
25
0 0 4 8 12 16 20 24 Time (hours)
Figure 2.8. Uptake or equilibrium curves for willow roots following the addition of radiolabeled compounds. Percent removal of radioactivity over time for root concentration factor studies (mean ± SE plotted, n = 3). The 16 hr day and 8 hr night
(shaded) are indicated. * Significantly different means between Roots versus
Roots+Wood (p < 0.05) from Table A2.7 ANOVA analysis.
69 Carmen G. Franks - Salix exigua Phytoremediation
1000
Sandbar willow log RCF = 0.93 log Kow – 1.10 Diltiazem RCF R2 = 1.00 100 g/mL)] g/mL)] μ barley g/g)/( μ
[( 10 poplar Root concentration factor (RCF) (RCF) factor Root concentration
17alpha- Atrazine Diltiazem Diazepam Ethynylestradiol 1 2.61 2.70 2.82 3.67 2.5 2.75 3 3.25 3.5 3.75
Log octanol-water partition coefficient (Kow)
Figure 2.9. Calculated and experimentally determined root concentration factor values
(mean ± SE) for Salix exigua. Values also calculated using compounds’ log octanol-water partition coefficients and the equations for barley roots (Briggs et al., 1982) and hybrid poplar roots (Burken and Schnoor, 1998). Diltiazem RCF value was not calculated using excised roots, but from whole plant uptake studies and the proportion of the compound within the root, therefore DTZ was excluded from this equation development.
70 Carmen G. Franks - Salix exigua Phytoremediation
100 Root Soluble Root Bound Root, Whole Wood, Whole 75
50 Recovery (%) Recovery
25
0 Excised Roots+ Excised Roots+ Excised Roots+ Roots Wood Roots Wood Roots Wood
17α-Ethynylestradiol Diazepam Atrazine
Figure 2.10. Distribution of recovered radioactivity within the roots and cuttings of root concentration factor experiment plants for Salix exigua. Mean ± SE are shown.
71 Carmen G. Franks - Salix exigua Phytoremediation
Estimated transpiration stream concentration factor (TSCF) values were not significantly
different when calculated using percent recovery or percent distribution of recovered (t-
test, ATZ, t = 0.520, df = 5, p = 0.625; DZP, t = 1.896, df = 5, p = 0.116; EE2, t = -2.287,
df = 5, p = 0.071), therefore the data shown in Figure 2.11 are calculated from percent recovery. DTZ values were not used in this equation development as there was no movement into the shoot. The equation developed from these calculations with an R2 =
0.822 for Salix exigua TSCF was (Figure 2.11):
2 log TSCF = 0.949 exp[-(log Kow – 2.90) / 0.321].
TSCF values for willow are less than 1 and decline in the order ATZ > DZP > EE2, with
ATZ’s value occurring between calculated TSCF values determined using Briggs et al.
(1982) and Burken and Schnoor’s (1998) equations and the compounds’ log Kow values
(Figure 2.11). The TSCF value for EE2 was just below the value calculated for barley roots. The experimentally derived TSCF value for DZP was approximately 1.5x greater than the calculated TSCF value.
72 Carmen G. Franks - Salix exigua Phytoremediation
1.00
0.75 log TSCF = 0.949 exp{-(log Kow – 2.90)2 / 0.321}
g/mL)] g/mL)] 2
μ R = 0.822 poplar 0.50 g/mL)/( μ Sandbar willow barley 0.25 (TSCF) [( 17alpha- Atrazine Diltiazem Diazepam Ethynylestradiol 0.00 2.61 2.70 2.82 Transpiration stream concentration factor factor concentration stream Transpiration 3.67 2.5 2.75 3 3.25 3.5 3.75 Log octanol-water partition coefficient (K ) ow
Figure 2.11. Calculated transpiration stream concentration factor values (mean ± SE) for
Salix exigua. Diltiazem was excluded from this equation development. Values also calculated using compounds’ log octanol-water partition coefficients and the equations for barley shoots (Briggs et al., 1982) and hybrid poplar shoots (Burken and Schnoor,
1998).
73 Carmen G. Franks - Salix exigua Phytoremediation
2.4 DISCUSSION
Uptake. Salix exigua, a common riparian plant, was examined for its ability to remove
trace levels of three pharmaceuticals: EE2, DZP, and DTZ, and the herbicide ATZ, from solution. Hydroponically grown Salix exigua was very effective at removing EE2, DZP and DTZ from solution with percent removal of 88, 56 and 77% removal, respectively, in
24 hours. The herbicide ATZ, was also removed with 50% taken up in the same conditions and interval.
The uptake curves for EE2, DTZ and DZP were similar, while ATZ had had more
gradual uptake (Figure 2.4). The final uptake values between compounds are generally
consistent with predicted behaviour based on the compound’s logarithm octanol-water partitioning coefficient (log Kow). In order from highest to lowest log Kow values the compounds were sequenced: EE2 (3.67) >> DZP (2.82) > DTZ (2.70) > ATZ (2.61). This order was largely reflected in uptake, thus the compounds were sequenced EE2 > DTZ >
DZP > ATZ (Figure 2.4).
A higher octanol-water partitioning coefficient would reflect a compound’s preference or
attraction for the organic phase versus water. Consequently, a larger log Kow value
suggests the compound would sorb to root cells at a faster rate than a compound with a
lower log Kow value. This principle is reflected in the distribution of the different
compounds. EE2, with its approximate 10-fold larger Kow, apparently had a greater
preference by the root cells than the other 3 compounds and this is reflected in the faster
uptake curve (Figures 2.4 and 2.6). EE2 uptake was expected to be correlated with root
74 Carmen G. Franks - Salix exigua Phytoremediation
weight given its binding preference to the roots (RCF > 100) and its log Kow value, but this was not consistently the case across the study period, likely due to the limited variation in root weights of the experimental plants.
The reversal of DTZ and DZP for uptake, may be suggesting processes are increasing the uptake of DTZ. As DTZ in this study was in the form DTZ-hydrochloride, it was not a neutral compound as the other 3 are expected to be, and would therefore become highly ionized, at external solution pH and physiological pH. DTZ uptake would then become a function of the proportion ionized and electrochemical gradients. ATZ uptake was expected to be correlated to shoot mass or transpired volumes given that it has a low log
Kow value and that its an herbicide with an action site in the shoot, and this was the case at t = 24.
Plant fresh weights and total volumes transpired did not vary significantly between compounds, except for wood weight, indicating unequal stem sizes across the treatments
(Table 2.1 and A2.1). Fortunately, the differences are unlikely to affect uptake significantly since the wood contained low levels of the compounds. The effects of wood weight would be expected to be negligible, unless uptake is correlated to transpiration, which it generally was not (Table A2.5). The similar fresh weights and transpiration volumes of the replicate plants among compounds allows for comparisons between the compounds for uptake and distribution, as uptake is primarily a function of root mass.
75 Carmen G. Franks - Salix exigua Phytoremediation
The lack of consistent correlation between uptake and plants’ fresh weights and volume
transpired (Table A2.4) is likely due to the limited range of willow weights used for the
replicates for each compound and across compounds. It would be expected that there
would be a positive relationship between plant weight and uptake, as well as transpiration and uptake if the compound moved freely into the transpiration stream. The relationship
between uptake and root mass/volume was positive, indicating that increasing the amount
of root area for sorption increased the uptake. As well, if a larger volume was moving
within the transpiration stream, there would be a larger carrier base for freely moving
compounds.
Cumulative transpired volumes were correlated with shoot mass for DZP and EE2 (Table
A2.5, Spearman’s rho, p < 0.05). For all compounds, cumulative transpired volumes were
correlated with total plant weight (Table A2.5, Spearman’s rho, p > 0.05). This
correlation is likely evident due to the sample sizes and the variance in plants’ weights
(Table 2.1).
No plants showed signs of phytotoxicity that would be evidenced by discoloration of the
leaves or alterations in transpiration (Trapp et al., 2000). Since ATZ is an herbicide, there
was the potential for an observable effect. As well, DTZ, the calcium-channel blocker,
might have had an effect if it was impeding the plant’s calcium transport.
Uptake and distribution of the herbicide ATZ in hybrid poplar trees have been studied
extensively by Burken and Schnoor (1996, 1997 and 1998). The findings of the present
76 Carmen G. Franks - Salix exigua Phytoremediation study correspond to those prior results with levels being highest in the shoots, followed by the cutting and then roots. Translocation from roots to shoots was apparently rapid when compared to the other compounds (Figures 2.6 and 2.7) and is also consistent with
Burken and Schnoor (1997).
Root uptakes for the root concentration factor (RCF) experiments were lower for excised roots than for roots attached to the cuttings among the compounds over the sampling period (Figure 2.8; Table A2.7, ANOVA). Differences in root uptake between excised roots and roots attached to the cutting may be explained by the travel of the compound into the cutting by diffusion or slight evaporation from the severed shoot node. It was expected that uptake would be equivalent between excised roots and roots attached to the cutting as the root masses were similar (root mass and root uptake have a positive relationship), but this appears to not be the case.
Examination into distribution of the compounds within the roots and wood in the RCF plants, in terms of soluble and bound fractions, returned low and somewhat variable recovery levels, especially for ATZ. Recoveries for EE2, DZP and ATZ were 54, 40 and
10%, respectively. Proportions within the roots generally reflected the root proportions observed from the whole plant uptake studies. This suggests that the proportions observed with whole plant uptake can be largely accounted for by this initial equilibration between root and external solution concentrations (represented as RCF).
77 Carmen G. Franks - Salix exigua Phytoremediation
Calculated RCF values using whole plant uptake data were compared to experimentally derived RCF values from the root uptake experiments and were consistent except for
ATZ (t-test, t = -3.25, df = 5, p = 0.023). Briggs et al. (1982) found that using the percentage of radioactivity within the roots from whole plant uptake studies to calculate
RCF resulted in values similar to those determined experimentally using excised roots.
Similar results may be represented here for willow. Based on the log Kow value for DTZ, it was expected to behave similarly to ATZ and DZP with some of the compound moving into the shoot, but this did not occur suggesting DTZ may not follow the same relationship.
The relationship between TSCF and log Kow is generally expressed as a bell shaped or normal curve, with maximum uptake corresponding to log Kow values around 2, although this value varies across plants. The log Kow values used in the present study ranged from
2.61 to 3.67, and therefore only the decreasing half of the normal curve is represented and the maximum uptake log Kow could not be determined for willow. All estimated TSCF values fell below 1 indicating the compounds are not easily taken up into the transpiration stream.
The studies for barley (Briggs et al., 1982) and hybrid poplar (Burken and Schnoor,
1998) involved multiple chemicals with a range of log Kow values enabling the determination of optimum uptake. Optimal uptake occurs with moderately hydrophobic compounds (log Kow = 1.0 – 3.5), with values greater than 3 being highly sorbed to roots
(Briggs et al., 1982; Burken and Schnoor, 1998). The range of log Kow values used in this
78 Carmen G. Franks - Salix exigua Phytoremediation
study was 2.61 – 3.67. Uptake from solution and movement into the transpiration stream
followed this suggested optimum range for uptake for willow.
A preliminary investigation into the uptake and time course distribution of EE2 with
willow was performed prior to the experiments discussed in this chapter and reported in
Appendix B. Three willow plants were treated with 3 concentrations of EE2 (0, 1 and 10
μg EE2 + 16,667 Bq 3H-EE2) in the same manner as the uptake studies described here
but the plants were harvested after 8 hours. Recovery was lower for 2 of the 3 plants from
the extraction procedures (soluble, bound and whole cutting oxidation), but distribution
of recovered radioactivity suggests concentration has minor influences on distribution
and uptake. As well, 3 plants were exposed to a single concentration of EE2 (0 μg EE2 +
16,667 Bq 3H-EE2) and one plant harvested at 8, 24 and 48 hours to discern any differences in distribution over time. There appears to be a change in distribution of EE2 within willow over time, with an increasing proportion becoming bound to the root (and
hence less remaining soluble within the root).
Conclusions. Salix exigua was effective at uptake of the three pharmaceuticals in 24
hours. The synthetic estrogen EE2 was removed most effectively, with most of the
compound becoming bound to the roots and little transport to shoots. DTZ, the anti-
hypertensive, was removed from solution at an intermediate rate and remained within the
soluble component within the root. The anti-anxiety drug DZP remained primarily soluble with over half of the compound taken up being transported to the shoots. The herbicide ATZ distribution was similar to that previously reported for poplars.
79 Carmen G. Franks - Salix exigua Phytoremediation
Predictive relationships between the pharmaceuticals’ physio-chemical property, the
octanol-water partitioning coefficient, and willow roots and shoots were developed and
found to be similar to previously reported relationships for other compounds for hybrid
poplar roots. However, DTZ uptake and distribution were different then predicted
indicating that aspects other than partitioning apply. Based on the results for EE2 and
DZP, it appears that pharmaceuticals behave similarly to other chemicals and this should
allow for predictable behaviour with respect to phytoremediation.
I thus conclude that phytoremediation appears to be a promising solution for remediating environmental pharmaceuticals in water. This treatment may be seasonally limited, however, and prospective phytoremediation would be far less effective during the winter season when willow and other plants are leafless and often dormant. A concern may also be the release of these compounds back into the environment with winter leaf drop or the decay of fallen plants. Sequestration of these compounds within a bound form may inhibit re-entry into the environment. The more potent and influential compound on aquatic life, the synthetic estrogen EE2, is also significantly bound within willow. The herbicide ATZ is most likely to return to the environment through leaf drop or plant decay as it remains primarily in the soluble form within the leaves. As well, the results observed here with willow are consistent with prior research with poplar, suggesting poplar research may be applicable to willow, and vice versa.
80 Carmen G. Franks - Salix exigua Phytoremediation
LITERATURE CITED
Briggs, GG, Bromilow, RH, Evans, AA. 1982. Relationships between lipophilicity and root uptake and translocation of non-ionised chemicals by barley. Pestic. Sci. 13: 495-504.
Burken, JG and Schnoor, JL. 1996. Phytoremediation: Plant uptake of atrazine and role of root exudates. J. Environ. Eng. 122: 958-963.
Burken, JG and Schnoor, JL. 1997. Uptake and metabolism of atrazine by poplar trees. Environ. Sci. Technol. 31: 1399-1406.
Burken, JG and Schnoor, JL. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32: 3379-3385.
Cargouet, M, Perdiz, D, Mouatassim-Souali, A, Tamisier-Karolak, S, Levi, Y. 2004. Assessment of river contamination by estrogenic compounds in Paris area (France). Sci. Total Environ. 324: 55-60.
Dietz, AC and Schnoor, JL. 2001. Advances in phytoremediation. Environ. Health Perspect. 109: 163-168.
Gibeaut, DM, Hulett, J, Cramer, GR, Seemann, JR. 1997. Maximal biomass of Arabidopsis thaliana using a simple, low-maintenance hydroponic method and favorable environmental conditions. Plant Physiol. 115: 317-319.
Gorshkov, S, Gorshkova, G, Colorni, B, Gordin, H. 2004. Effects of natural estradiol-17 beta and synthetic 17 alpha-ethynylestradiol on direct feminization of European sea bass Dicentrarchus labrax. J. World Aquacult. Soc. 35: 167-177.
Kolpin, DW, Furlong, ET, Meyer, MT, Thurman, EM, Zaugg, SD, Barber, LB, Buxton, HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci. Technol. 36: 1202-1211.
Kuch, HM and Ballschmiter, K. 2001. Determination of endocrine-disrupting phenolic compounds and estrogens in surface and drinking water by HRGC-(NCI)-MS in the picogram per liter range. Envion. Sci. Technol. 35: 3201-3206.
Orchard, BJ, Doucette, WJ, Chard, JK, Bugbee, B. 2000. Uptake of trichloroethylene by hybrid poplar trees grown hydroponically in flow-through plant growth chambers. Environ. Toxicol. Chem. 19: 895-903.
Paterson, S, Mackay, D, McFarlane, C. 1994. A model of organic chemical uptake by plants from soil and the atmosphere. Environ. Sci. Technol. 28: 2259-2266.
81 Carmen G. Franks - Salix exigua Phytoremediation
Schnoor, J. 2002. GWRTAC Technology Evaluation Report. Phytoremediation of Soil and Groundwater. Iowa City: The University of Iowa. Report nr E-Series: TE-02- 01. 45 p. Available online: www.gwrtac.org
Sengl, M and Krezmer, S. 2003. Proficiency tests for pharmaceuticals in different waters. Accred. Qual. Assur. 8: 523-529.
Servos, MR, Bennie, DT, Burnison, BK, Jurkovic, A, McInnis, R, Neheli, T, Schnell, A, Seto, P, Smyth, SA, Ternes, TA. 2005. Distribution of estrogens, 17β-estradiol and estrone, in Canadian municipal wastewater treatment plants. Sci. Total Environ. 336: 155-170.
Shone, MGT and Wood, AV. 1974. A comparison of the uptake and translocation of some organic herbicides and a systemic fungicide by barley. I. and II. J. Exp. Bot. 25: 390-400.
Sosiak, A and Hebben, T. 2005. A preliminary survey of pharmaceuticals and endocrine disrupting compounds in treated municipal wastewaters and receiving waters of Alberta. Environmental Monitoring and Evaluation Branch. Alberta Environment. Pub. No: T/773. 64 p. Available online: http://www3.gov.ab.ca/env/info/infocentre/publist.cfm
Stan, HJ and Linkerhägner, M. 1992. Vom Wasser 79: 75-88.
Ternes, TA, Kreckel, P, Mueller, J. 1999. Behaviour and occurrence of estrogens in municipal sewage treatment plans – II. Aerobic batch experiments with activated sludge. Sci. Total Environ. 225: 91-99.
Trapp, S, Zambrano, KC, Kusk, KO, Karlson, U. 2000. A phytotoxicity test using transpiration of willows. Arch. Environ. Contam. Toxicol. 39: 154-160.
Werner, J, Wautier, K, Evans, RE, Baron, CL, Kidd, K, Palace, V. 2003. Waterborne ethynylestradiol induces vitellogenin and alters metallothionein expression in lake trout (Salvelinus namaycush). Aquat. Toxicol. 62: 321-328.
82 Carmen G. Franks - Arabidopsis Phytoremediation
CHAPTER 3 Phytoremediation of trace levels of pharmaceuticals, diltiazem,
diazepam and 17α-ethynylestradiol, and the herbicide atrazine, with Arabidopsis
thaliana
3.1 INTRODUCTION
This present investigation studied the potential capability of Arabidopsis thaliana to
uptake and transport three pharmaceuticals that represent common classes of substances
that have been detected in municipal wastewater, including diltiazem (DTZ), a calcium
channel blocker, diazepam (Valium®) (DZP), an anti-anxiety drug; and 17α- ethynylestradiol (EE2), a synthetic birth control hormone (a common component of the contraceptive pill). Although Arabidopsis is not a practical field plant for
phytoremediation, its genome has been sequenced and it can provide useful insight into the genes involved in uptake, translocation and metabolism of pharmaceuticals within other plants (Cobbett and Meagher, 2002).
This study of uptake and distribution of these compounds within Arabidopsis provides a
comparison to Salix exigua’s uptake and distribution of these compounds. As well, a
second plant can help to confirm the behaviour of pharmaceuticals within plants and the
properties that guide their uptake and distribution.
Phytoremediation of pharmaceuticals is not widely published in the literature, but is a
new direction for research that should be considered, particularly as the quantities of
detectable contaminants are only increasing as technology and methodologies change.
83 Carmen G. Franks - Arabidopsis Phytoremediation
3.2 MATERIALS AND METHODS
3.2.1 Chemicals
The pharmaceuticals and other chemicals used for this study are described in Chapter 2.
3.2.2 Plants, hydroponics and treatments
Arabidopsis thaliana (var. Columbia) was started from seed and seedlings were grown using a hydroponic setup in a controlled environment growth chamber (Figure 3.1). The hydroponic system was described in Chapter 2, with adaptation for Arabidopsis. To provide a germinating and support medium for Arabidopsis seeds, horticultural Rock
Wool made from basalt fibers (FibrGrow Horticultural Products, Sarnia, Ontario), was cut into 1.5 x 3.0 cm cubes and inserted into the drilled holes. Toothpicks supported the
Rock Wool above the water surface allowing submersion about 2 cm into solution. Two seeds were placed on the top of each exposed end of Rock Wool for later thinning to one plant. Water level was kept between 1.5 to 2 cm below seed level to promote germination and proper growth, as was suggested by Gibeaut et al. (1997). Growth chamber and chamber conditions for Arabidopsis were the same as described in Chapter 2.
Arabidopsis plants approximately 4 weeks old and approaching the flowering stage were used for the uptake studies.
84 Carmen G. Franks - Arabidopsis Phytoremediation
Figure 3.1. Hydroponic system for Arabidopsis seed germination and growth.
85 Carmen G. Franks - Arabidopsis Phytoremediation
3.2.3 Uptake studies
The uptake study set-up is described in Chapter 2. The uptake experiments were replicated twice with similar results and one data set is presented. The second data set not presented involved n = 4 for each compound.
Similar sized plants were chosen and inserted into the culture tubes. Plant shoots were
supported so that only the roots were submerged. At the time of harvest (24 hr) the plants
were removed from solution, the roots were rinsed in nutrient solution to remove surface
solution on the roots, and divided into shoots and roots. Each was weighed fresh and stored at -20 °C until analysis. For each compound, replicate plants numbered 10, 6, 6 and 5 for EE2, DTZ, DZP and ATZ for the data set presented, respectively. Transpiration volumes were monitored as described in Chapter 2.
3.2.4 Soluble fractions
A methanol solvent extraction procedure is described in Chapter 2 and was carried out on
some of the replicate plants. For EE2, n = 4; DTZ, n = 4; DZP, n = 5; ATZ, n = 3.
However, due to initial problematic experimental procedures using SPE to remove
pigments from the shoot extracts (see Appendix B for some results and method changes),
the numbers of results considered reliable decreased. Numbers of replicates from solvent
extraction that were reliable became: EE2, n = 4; DTZ, n = 2; DZP, n = 4, ATZ, n = 1.
86 Carmen G. Franks - Arabidopsis Phytoremediation
3.2.5 Bound fractions
Quantification of bound radioactivity that may have been within the cell walls that was
not removed with organic solvent was performed on Arabidopsis. Residues that had
undergone solvent extraction were oxidized to extract bound tritium or carbon-14, as
described in Chapter 2. Some whole replicate plants that had not undergone solvent extraction underwent oxidation to confirm levels of radioactivity being recovered through combination of the solvent extraction and oxidation of residues. For whole plants that underwent oxidation: DTZ, n = 2; DZP, n = 1; EE2, n = 6; ATZ, n = 2.
3.2.6 Root concentration factor and transpiration stream concentration factor
The root concentration factors (RCF) for the four compounds and Arabidopsis were
estimated using results from the whole plant uptake studies and the equation (Shone and
Wood, 1974):
root uptake = RCF (μg/mL) x root mass (g) x external solution concentration
(μg/mL).
RCF values were also calculated for the four compounds using log Kow values and the equations developed by Briggs et al. (1982) for barley roots:
log (RCF - 0.82) = 0.77 log Kow - 1.52,
and by Burken and Schnoor (1998), for hybrid poplar roots:
87 Carmen G. Franks - Arabidopsis Phytoremediation
log (RCF - 3.0) = 0.65 log Kow – 1.57.
Transpiration stream concentration factor (TSCF) estimates for the four compounds in
Arabidopsis were estimated using results from whole plant uptake studies and the
equation (Briggs et al., 1982):
TSCF = [concentration in the shoot (μg) / volume transpired (mL)] / [(external solution concentration (μg/mL)initial + external solution concentration (μg/mL)final )/
2].
Although transpiration may not have been continuous, the means of initial and final external solution concentrations were used to estimate overall TSCF (Briggs et al., 1982).
TSCF were also calculated for the four compounds using log Kow values and the
equations developed by Briggs et al. (1982) for barley shoots:
2 log TSCF = 0.784 exp[-(log Kow – 1.78) / 2.44],
and by Burken and Schnoor (1998), for hybrid poplar shoots:
2 log TSCF = 0.756 exp[-(log Kow -2.50) / 2.58].
88 Carmen G. Franks - Arabidopsis Phytoremediation
3.3 RESULTS
3.3.1 Uptake studies time course
Arabidopsis was capable of removing EE2, DZP and DTZ with removal at 85, 59 and
57%, respectively, in 24 hours. The herbicide ATZ was employed as a positive control as
it is known to be readily taken up by plants. Arabidopsis removed 52% of ATZ within the
24 hour period.
The percentage uptake, or removal from solution for DTZ, ATZ, and DZP over time
followed a very similar pattern and maximum values at 24 hours (Figure 3.2). Rapid
uptake within the first 2 hours was consistent among the compounds, reaching
approximately 25% removal from solution for DZP, DTZ and ATZ. Arabidopsis removed 74% of the synthetic estrogen within the first 2 hours.
The order of highest to lowest uptake amount is a reflection of the compounds’ log Kow:
EE2 (85% uptake, log Kow 3.67) >> DZP (59% uptake, log Kow 2.82) > DTZ (57%
uptake, log Kow 2.70) > ATZ (52% uptake, log Kow 2.61).
Among compounds, the plants’ weights (root, shoot and total weight) and total volume
transpired did not vary significantly (Table 3.1 and Table A3.1, ANOVA, p > 0.05), thus,
replicate plants were of similar size and apparently similar health.
No plants showed signs of phytotoxicity during the period of study and with the
concentrations used. There was no observable discoloration of the leaves or significant
89 Carmen G. Franks - Arabidopsis Phytoremediation alteration in transpiration rates, which were approximated by the straight lines of cumulative transpiration up to 10 hours (Figure 3.3). A slight reduction in transpiration rate between 10 and 24 hours probably reflects reduced transpiration during the night phase of the study.
Similar cumulative transpiration trends over the 24 hour period were observed for all replicate plants for all compounds (Figure 3.3 and Table A3.2, ANOVA). Transpiration volume at t = 2 varied significantly between all compounds except between DZP or DTZ and EE2 but did vary between DTZ and EE2 (Table A3.3, LSD, p < 0.05). At t = 4 transpiration volume varied significantly between EE2 and all other compounds but not between ATZ, DTZ and DZP (Table A3.3, LSD, p < 0.05).
A correlation analysis (Table A3.4, Spearman’s rho) for uptake with fresh weight and volume transpired at each sampling time, found no significant correlation between these factors and uptake for DTZ or DZP (Spearman’s rho, p > 0.05). There was no significant correlation between uptake and volume transpired for any of the compounds (Spearman’s rho, p > 0.05). ATZ uptake was correlated with total fresh weight at time t = 10 and t =
24 (Spearman’s rho, p = 0.037 and p = 0.037). EE2 uptake was significantly correlated with root, shoot and total fresh weight at t = 2 (Spearman’s rho, p < 0.05). Cumulative volume transpired over time was not consistently correlated to plant fresh weights (Table
A3.5, Spearman’s rho, p > 0.05), except for EE2 (p < 0.05).
90 Carmen G. Franks - Arabidopsis Phytoremediation
Volatility of the compounds under the growth chamber conditions was negligible for the four compounds in the plant-free control conditions. Similarly, testing for compound binding to the culture tubes revealed no remaining glassware-bound radioactivity up to the t = 24 sampling time.
91 Carmen G. Franks - Arabidopsis Phytoremediation
100
75
50
25 17α-Ethynylestradiol
0 100
75
50
25 Diltiazem
0 100 Uptake (%)
75
50
25 Diazepam
0 100
75
50
25
Atrazine 0 0 4 8 12 16 20 24 Time (hours) Figure 3.2. Uptake from solution of 0.04 μg/ mL of 17α-ethynylestradiol (n = 10), diazepam (n = 6), diltiazem (n = 6), and atrazine (n = 5), by Arabidopsis. Mean ± SE are plotted for the 24 hour period. The16 hr day and 8 hr night (shaded) are indicated.
92 Carmen G. Franks - Arabidopsis Phytoremediation
Table 3.1. Summary information for Arabidopsis plants’ fresh weight and total volume transpired in the 24 hour study. Mean ± SE are shown.
Root Shoot Total Volume n fr.wt. (g) fr.wt. (g) Transpired (mL)
17α-Ethynylestradiol 10 0.79 ± 0.12 1.85 ± 0.25 9.65 ± 0.81
Diltiazem 6 0.73 ± 0.10 1.65 ± 0.21 9.13 ± 0.97
Diazepam 6 0.70 ± 0.10 1.47 ± 0.21 8.83 ± 0.90
Atrazine 5 0.67 ± 0.18 1.75 ± 0.17 10.85 ± 1.40
93 Carmen G. Franks - Arabidopsis Phytoremediation
12
8
4
17α-Ethynylestradiol 0 12
8
) mL ( 4 me u Diltiazem 0 ired vol
p 12 trans e 8 Cumulativ 4
Diazepam 0 12
8
4
Atrazine 0 04812162024 Time (hours) Figure 3.3. Cumulative transpired volumes measured for Arabidopsis at each sampling time for the four compounds. Mean ± SE are shown. The 16 hr day and 8 hr night
(shaded) are indicated.
94 Carmen G. Franks - Arabidopsis Phytoremediation
3.3.2 Distribution
Plant components underwent solvent extraction to retrieve the soluble fraction, and the
remaining residue was oxidized to analyze the bound fraction. Recovery levels varied
across the compounds. Final percent recoveries for EE2, DTZ, DZP and ATZ were 90,
55, 35, and 38%, respectively (Figures 3.4 – 3.9). Values for root and shoot are combined
extraction methods; soluble, bound and whole plant oxidation. Note that the value of n for ATZ, DTZ, and DZP have changed from those shown in Figures 3.2 and 3.3 and
Tables 3.1, A3.1 – A3.5, as the original methods resulted in unreliable values for recovery for some of the replicate plants (See Appendix B for discussion on methods changes).
Distribution ratios for Arabidopsis, root:shoot, from combined results (soluble solvent extract, bound residue oxidation and whole plant oxidation) were: EE2, 19:1; DTZ, 4:3
(1.3:1); DZP, 1:10; and ATZ, 1:10. Separate results for root:shoot ratios were EE2,
15.5:1 (soluble + bound) and 20:1 (oxidized whole plant); DTZ, 2:1 (soluble + bound) and 1:1.4 (oxidized whole plant); DZP, 1.2 (soluble + bound) and 1:2.5 (oxidized whole plant); ATZ, 1:9 (soluble + bound) and 1:9 (oxidized whole plant).
As recovery was varied and relatively low for the 4 compounds, but consistency with
distribution of recovered radioactivity among replicates was evident, the data is presented
both ways; as percentage recovery and as distribution of recovered radioactivity (Figures
3.6 – 3.9). The consistency within distribution of recovered radioactivity among
replicates allows for discerning of trends that might otherwise not be as evident.
95 Carmen G. Franks - Arabidopsis Phytoremediation
EE2 had the highest recovery levels and most consistent proportion distribution of
recovered (due to high recovery). EE2 became largely bound within the root, with greater
than 80%, with a small portion remaining soluble in the root. Shoot soluble fractions did
occur and were approximate in proportion to shoot bound fractions, making up less than
10% of recovered radioactivity (Figures 3.5 and 3.6). DTZ remained primarily soluble
within the root, making up approximately 50% of recovered radioactivity. A similar
proportion moved into the shoots and also remained soluble as less than 45% of recovered. Root and shoot bound fractions were comparatively small, but did occur. The highest degree of variability in recovery was observed with DTZ, which extended into the proportion distribution of recovered (Figures 3.5 and 3.7).
DZP was found in the largest proportion in the shoot as a soluble form, greater than 60%
of recovered radioactivity. Bound fractions in both the root and shoot did occur, but in
comparatively small proportions. The root soluble fraction made up the second largest
fraction with less than 40% (Figures 3.5 and 3.8). ATZ did not become extensively bound
to root or shoot components, with the largest portion remaining soluble in the shoots,
making up over 90% of the recovered radioactivity (Figures 3.5 and 3.9).
96 Carmen G. Franks - Arabidopsis Phytoremediation
100
Root Shoot Total Plant
50 Recovery (%) Recovery
0 17alpha- Diltiazem Diazepam Atrazine Ethynylestradiol
Figure 3.4. Distribution of recovered radioactivity from oxidation of whole root and shoot and total plant recovery for radiolabeled compound added to Arabidopsis (mean ±
SE). For 17α-ethynylestradiol Root and Shoot, n = 6; diltiazem, n = 2; diazepam, n = 1; atrazine, n = 2. For Total Plant, 17α-ethynylestradiol, n = 10; diltiazem, n = 4; diazepam, n = 5; atrazine, n = 3.
97 Carmen G. Franks - Arabidopsis Phytoremediation
100
Root Soluble Root Bound Shoot Soluble Shoot Bound
50 Recovery (%) Recovery
0 17alpha- Diltiazem Diazepam Atrazine Ethynylestradiol
Figure 3.5. Distribution of recovered radioactivity among soluble and bound fractions in
roots and shoots for radiolabeled compound added to Arabidopsis (mean ± SE). For 17α- ethynylestradiol, n = 4 soluble + bound; diltiazem, n = 2 soluble + bound; diazepam, n =
4 soluble + bound; atrazine, n = 1 soluble + bound.
98 Carmen G. Franks - Arabidopsis Phytoremediation
Root, Whole Shoot, Whole Root Soluble Shoot Soluble Root Bound Shoot Bound
100 A
50 Recovery (%) Recovery
0 100% B red radioactivity (%) radioactivity red 50%
0% recove of Distribution
Figure 3.6. Arabidopsis individual replicate plant percent recovery of radioactivity for the synthetic estrogen 17α-ethynylestradiol (n = 10). A, actual percent recovered radioactivity for root and shoot components; B, percent distribution of recovered radioactivity from A. Note y-axis exceeds 100%.
99 Carmen G. Franks - Arabidopsis Phytoremediation
Root, Whole Shoot, Whole Root Soluble Shoot Soluble Root Bound Shoot Bound
100 A
50 Recovery (%) Recovery
0 100% B red radioactivity (%) radioactivity red 50%
0% Distribution of recove of Distribution
Figure 3.7. Arabidopsis individual replicate plant percent recovery of radioactivity for the anti-hypertensive diltiazem (n = 4). A, actual percent recovered radioactivity for root and shoot components; B, percent distribution of recovered radioactivity from A.
100 Carmen G. Franks - Arabidopsis Phytoremediation
Root, Whole Shoot, Whole Root Soluble Shoot Soluble Root Bound Shoot Bound
100 A
50 Recovery (%)
0 100% B red radioactivity (%) radioactivity red 50%
Distribution of recove of Distribution 0%
Figure 3.8. Arabidopsis individual replicate plant percent recovery of radioactivity for the anti-anxiety drug diazepam (n = 5). A, actual percent recovered radioactivity for root and shoot components; B, percent distribution of recovered radioactivity from A.
101 Carmen G. Franks - Arabidopsis Phytoremediation
Root, Whole Shoot, Whole Root Soluble Shoot Soluble Root Bound Shoot Bound
100 A
50 Recovery (%) Recovery
0 100% B
red radioactivity (%) radioactivity red 50%
0% Distribution of recove of Distribution
Figure 3.9. Arabidopsis individual replicate plant percent recovery of radioactivity for the herbicide atrazine (n = 3). A, actual percent recovered radioactivity for root and shoot components; B, percent distribution of recovered radioactivity from A.
102 Carmen G. Franks - Arabidopsis Phytoremediation
3.3.3 Root concentration factor and transpiration stream concentration factor
Root concentration factors (RCF) for the four compounds and Arabidopsis were
estimated using results from the whole plant uptake studies and the equation (Shone and
Wood, 1974):
root uptake = RCF (μg/mL) x root mass (g) x external solution concentration
(μg/mL).
Estimated RCF values were not significantly different when calculated using percent
recovery or percent distribution of recovered (t-test, p>0.05), therefore RCF was
calculated using percent recovery (t-tests RCF, EE2, t = - 0.663, df = 9, p = 0.524; DZP, t
= -2.269, df = 4, p = 0.086; DTZ, t = -0.898, df = 4, p =0.420; ATZ, t = -2.174, df = 4, p
= 0.095).
The equation developed from these calculations and using percent recovery, with an R2 =
0.802, for Arabidopsis RCF was (Figure 3.10):
log (RCF) = 1.38 log Kow – 2.71.
Also presented are the RCF values obtained from Briggs et al. (1982) and Burken and
Schnoor (1998) equations.
103 Carmen G. Franks - Arabidopsis Phytoremediation
1000
log RCF = 1.38 log Kow – 2.71 100 R2 = 0.802 Arabidopsis g/mL)] μ
barley g/g)/( μ
10
(RCF) [( (RCF) poplar Root concentration factor factor Root concentration
2.61 2.70 2.82 3.67 1 Atrazine Diltiazem Diazepam 2.5 2.75 3 3.25 3.5 17alpha-3.75 Ethynylestradiol Log Kow Log octanol-water partition coefficient (Kow)
Figure 3.10. Calculated root concentration factor values (mean ± SE) for Arabidopsis.
Values also calculated using compounds’ log octanol-water partition coefficients and the equations for barley roots (Briggs et al., 1982) and hybrid poplar roots (Burken and
Schnoor, 1998).
104 Carmen G. Franks - Arabidopsis Phytoremediation
Transpiration stream concentration factors (TSCF) for the four compounds in
Arabidopsis were estimated using results from whole plant uptake studies and the
equation (Briggs et al., 1982):
TSCF = [concentration in the shoot (μg) / volume transpired (mL)) / [(external solution concentration (μg/mL)initial + external solution concentration (μg/mL)final )/
2].
Estimated TSCF values were not significantly different when calculated using percent recovery or percent distribution of recovered (t-test, p>0.05), therefore TSCF values were calculated using percent recovery (t-tests TSCF, EE2, t = 1.367, df = 9, p = 0.205; DZP, t
= -2.450, df = 5, p = 0.058; DTZ, t = -0.862, df = 4, p = 0.437; ATZ, t = -1.365, df = 4, p
= 0.244).
The equation developed from these calculations using percent recovery, with an R2 =
0.997, for Arabidopsis TSCF was (Figure 3.11):
2 log TSCF = 0.966 exp[-(log Kow – 2.63) / 0.694].
Also presented are the TSCF values calculated using the equations developed by Briggs
et al. (1982) and Burken and Schnoor (1998).
105 Carmen G. Franks - Arabidopsis Phytoremediation
1.25
1.00
log TSCF = 0.966 exp{-(log Kow – 2.63)2 / 0.694}
g/mL)] g/mL)] R2 = 0.997 μ 0.75
poplar g/mL)/(
μ 0.50
Arabidopsis barley
(TSCF) [( 0.25
2.61 2.70 2.82 3.67 Transpiration stream concentration factor factor concentration stream Transpiration 0.00 Atrazine Diltiazem Diazepam 2.5 2.75 3 3.25 3.5 17alpha- 3.75 Log Kow Ethynylestradiol Log octanol-water partition coefficient (Kow)
Figure 3.11. Calculated transpiration stream concentration factor values (mean ± SE) for
Arabidopsis. Values also calculated using compounds’ log octanol-water partition coefficients and the equations for barley shoots (Briggs et al., 1982) and hybrid poplar shoots (Burken and Schnoor, 1998).
106 Carmen G. Franks - Arabidopsis Phytoremediation
3.4 DISCUSSION
Uptake. Arabidopsis thaliana, the international model plant for scientific research, was
examined for its ability to remove trace levels of three pharmaceuticals, EE2, DZP, and
DTZ, and the herbicide ATZ, from solution. Arabidopsis was effective at removing EE2,
DZP and DTZ in a period of 24 hours with removal at 85, 59 and 57%, respectively. The
herbicide was employed as a positive control as it is known to be readily taken up by
plants. Arabidopsis removed 52% of ATZ within the 24 hour period.
Comparisons among compounds are possible due to the similar fresh weights and
transpiration volumes among replicate plants across compounds (Table 3.1 and A3.1), as
uptake is primarily a function of root mass.
Uptake curves for DTZ, ATZ, and DZP follow a very similar pattern and maximum
values at 24 hours (Figure 3.2). EE2 was removed from solution more quickly and at a
greater amount than the other 3 compounds (Figure 3.2). Similar uptake observed
between DZP, DTZ and ATZ is likely a function of the compounds similar log Kow
values (with a range of 2.61 – 2.82) and the rates of transpiration, weights of plants
(Table 3.1 and A3.1) and lipid content of the plants used. EE2 has a log Kow of 3.67
which correlates to a greater uptake amount after 24 hours (Figure 3.2). The rate of
uptake for EE2 may have slowed due to the decrease in external solution concentration to near 25% remaining within the first 2 hours, effectively reducing the available compound to a minimum. As well, hormones and hormone mimics are actively taken up from solution by plants (Geuns, 1978; Bhattacharya and Gupta, 1981; Hayat et al., 2001;
107 Carmen G. Franks - Arabidopsis Phytoremediation
Janeczko et al., 2003), suggesting the efficient uptake observed for EE2 could partially be due to a role of active uptake.
Comparing final uptake values between compounds, they are consistent with predicted behaviour based on the compound’s logarithm octanol-water partitioning coefficient (Log
Kow). In order of highest to lowest log Kow values and greatest to lowest uptake: EE2 >>
DZP > DTZ > ATZ. Generally, a larger octanol-water partitioning coefficient would reflect a compound’s preference or attraction for an organic phase over water. Therefore, a larger log Kow value would suggest the compound is sorbing to root cells at a faster rate than a compound with a lower log Kow value. This process is reflected in the distribution of the different compounds; EE2, with its approximate 10-fold larger Kow, has a greater preference to the root cells than the other 3 compounds and is reflected in the uptake curve (Figure 3.2 and 3.6).
The lack of correlation between plant weights, transpiration volumes and uptake is likely due to the limited range of Arabidopsis weights used for the replicates for each compound and across compounds. It would be expected that there would be a positive relationship between plant weight and uptake, as well as transpiration and uptake if the compound moved relatively freely into the transpiration stream. The relationship between uptake and plant mass/volume is a positive one, as increasing the amount of root area to sorb to is expected to increase the amount taken up. As well, if a larger volume was moving within the transpiration stream, it would suggest a larger carrier base for freely moving compounds to the shoot.
108 Carmen G. Franks - Arabidopsis Phytoremediation
Transpiration is expected to be correlated to shoot mass (Ray and Sinclair, 1998), but no
correlation between transpired volume and plant weight (Table A3.5; Spearman’s rho, p
> 0.05) was observed for these experiments except for EE2 (Table A3.5; Spearman’s rho,
p < 0.05), which is likely a function of a larger sample size (n = 10) and a wider range of
weights across replicates than for the other compounds (Table 3.1).
No plants showed signs of phytotoxicity, during the period of study and at the
concentrations used, as measured by discoloration of the leaves or significant alteration in
transpiration rates (Trapp et al., 2000) (Figure 3.3, Tables 3.1, A3.1 and A3.2). As ATZ is
an herbicide, there is the potential for an observable effect, but none were evident for the
period of the study. As well, DTZ, the calcium-channel blocker, could have had an effect
if it was acting upon the plant’s calcium-channels.
Distribution ratios for Arabidopsis, root:shoot, from combined results (soluble solvent extract, bound residue oxidation and whole plant oxidation) were EE2, 19:1; DTZ, 4:3
(1.3:1); DZP, 1:10; and ATZ, 1:10. Separate results for root:shoot ratios, comparing oxidized whole plant and soluble + bound fractions were very similar to the combined results listed above.
Comparing the proportions of distributed radioactivity across compounds, they again
follow the order of the compounds’ log Kow values, with the highest recovered proportion
in the root coinciding with the largest log Kow and vice versa, except for DTZ and DZP
109 Carmen G. Franks - Arabidopsis Phytoremediation
which are switched in their expected order for proportions within the root (Figure 3.4).
Movement into the shoot followed the order of the compounds’ log Kow values with the lowest values coinciding with the largest proportion of recovered radioactivity within the
shoot. ATZ is designed as an herbicide that acts on the photosynthetic chain within
pigmented plant cells, therefore it would be expected to be transported to the shoots,
which it was (Figures 3.5 and 3.6). The synthetic estrogen EE2 was strongly sorbed to the
roots, and taken up at a greater amount than the other compounds (Figure 3.3, 3.5 and
3.9) with the potential for some of this uptake to be accounted for active uptake of
hormones by plants (Geuns, 1978; Bhattacharya and Gupta, 1981; Hayat et al., 2001;
Janeczko et al., 2003). DTZ and DZP did not appear to induce any observable effects and
behaved as predicted from their log Kow values for both uptake and distribution.
Examination of the distribution of total recovered radioactivity shows consistency across
replicates, suggesting the oxidation process is variable in terms of recovery, but it is not
understood why. Each replicate shows consistent proportions in distribution likely due to
their oxidation in sequence, i.e. replicate A’s root and shoot were oxidized before moving
onto the next replicate, possibly allowing for the same margin of error in each replicate,
but not across the replicates. Oxidizer may not be the sole problem of poor recoveries,
but it is not understood what other factors are contributing, particularly when EE2
recoveries were 90% and the others were below 55%.
Calculations of RCF and TSCF values for Arabidopsis derived using the percentage of radioactivity within the roots and the shoots from whole plant uptake studies. This
110 Carmen G. Franks - Arabidopsis Phytoremediation
proposed method was suggested by Briggs et al. (1982) and found to result in values similar to those determined experimentally when using excised roots. Similar results may
be represented here for Arabidopsis. RCF and TSCF are typically calculated taking into
account degradation of the compound (Thompson et al., 1998), but as this was not
determined for these studies RCF and TSCF were calculated assuming no degradation of
the compound with all recovered radioactivity in parent form.
For comparison, RCF values and TSCF values were calculated two ways. One method
was using the percentage recovery within the roots and shoots, including soluble and
bound fractions. And the other method: as percent recovery was relatively low but
percent distribution of recovered radioactivity was consistent across replicates, an assumption was made that the percent distribution of recovered radioactivity was accurate. The total uptake radioactivity was therefore multiplied by the fraction of radioactivity expected to be within the root or shoot based on the root and shoot distribution fraction of recovered radioactivity.
Estimated RCF and TSCF values were not significantly different when calculated using
percent recovery or percent distribution of recovered (t-test, p>0.05), therefore RCF and
TSCF values were calculated from percent recovery. Determined equations for
Arabidopsis are:
log RCF = 1.38 log Kow – 2.71,
111 Carmen G. Franks - Arabidopsis Phytoremediation
2 and log TSCF = 0.966 exp[- (log Kow – 2.63) / 0.694].
Arabidopsis RCF values followed a similar trend as was found for barley roots (Briggs et
al., 1982) and hybrid poplar roots (Burken and Schnoor, 1998), but were at a steeper
slope (Figure 3.10). EE2 RCF values were 10x greater than those predicted using
published equations. The trend was similar to other documented relationships for other
compounds, therefore neutral pharmaceuticals may follow the same relationships as has been found for other compounds based on their log Kow values.
TSCF values followed a similar trend as other documented relationships, but were
approximately 1.2 - 1.5x larger for ATZ, DTZ and DZP with values nearing 1. EE2
calculated TSCF was similar to those predicted by Briggs et al. (1982) and Burken and
Schnoor’s (1998) equations (Figure 3.11). TSCF values of 1 imply passive uptake
following the transpiration stream. Values < 1 suggest the compounds are not easily
moved into the stream, and values > 1 suggest active uptake into the transpiration stream,
such as for nutrients K, P, and N (Orchard et al., 2000; Dietz and Schnoor, 2001). The
relationship between TSCF and log Kow is expressed as a bell shaped curve, with
maximum uptake corresponding to log Kow values around 2, but this value varies with the
plant of study. The log Kow values used in this experiment ranged from 2.61 to 3.67, and
therefore only the decreasing half of the bell is represented and the maximum uptake log
Kow could not be discerned for Arabidopsis.
112 Carmen G. Franks - Arabidopsis Phytoremediation
The studies for barley (Briggs et al., 1982) and hybrid poplar (Burken and Schnoor,
1998) involved multiple chemicals with a range of log Kow values enabling the determination of optimum uptake. Optimum uptake is suggested to occur with moderately hydrophobic compounds (log Kow = 1.0 – 3.5) (Briggs et al., 1982; Burken and Schnoor, 1998), with values greater than 3 being highly sorbed to roots. The range of log Kow values used in this study was 2.61 – 3.67. Uptake from solution and movement into the transpiration stream followed this suggested optimum range for uptake for
Arabidopsis. EE2, with a log Kow greater than 3, was highly sorbed to roots.
Conclusions. Arabidopsis was effective at uptake of 3 pharmaceuticals EE2, DZP and
DTZ from solution in a period of 24 hours, with percent removal of 85, 59 and 57%, respectively. The synthetic estrogen EE2 was removed at the greatest amount, with most of the compound becoming bound to the roots and little transport to shoots, during the 24 hour period. The anti-anxiety drug DZP was removed from solution at the second greatest amount remained primarily soluble with over half of the taken up compound being transported to the shoots. DTZ, the anti-hypertensive, and also remained primarily soluble with over half recovered within the root. The positive control, ATZ, was also taken up by Arabidopsis, with removal at 52% and distributed mostly to the shoots as a soluble fraction, with minimal binding within roots or shoots. As many enzymes and genes of Arabidopsis are known, the potential exists for predicting the degradation pathways within plants.
113 Carmen G. Franks - Arabidopsis Phytoremediation
Predictive relationships between pharmaceuticals’ physio-chemical property, octanol- water partitioning coefficient, and Arabidopsis roots and shoots were developed and found to be similar to previously reported relationships for other compounds and hybrid poplar and barley roots. This suggests that pharmaceuticals behave as other chemicals, allowing for predictive behaviour in the environment and in plants.
114 Carmen G. Franks - Arabidopsis Phytoremediation
LITERATURE CITED
Bhattacharya, B and Gupta, K. 1981. Steroid hormone effects on growth and apical dominance of sunflower. Phytochemistry 20: 989-991.
Briggs, GG, Bromilow, RH, Evans, AA. 1982. Relationships between lipophilicity and root uptake and translocation of non-ionised chemicals by barley. Pestic. Sci. 13: 495-504.
Burken, JG and Schnoor, JL. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32: 3379-3385.
Cobbett, CS and Meagher, RB. 2002. Arabidopsis and the Genetic Potential for the Phytoremediation of Toxic Elemental and Organic Pollutants. The Arabidopsis Book, eds. C.R. Somerville and E.M. Meyerowitz, American Society of Plant Biologists, Rockville, MD, doi/10.1199/tab.0032 http://www.aspb.org/publications/Arabidopsis/ - this publication is only available as an on-line text.
Dietz, AC and Schnoor, JL. 2001. Advances in phytoremediation. Environ. Health Perspect. 109: 163-168.
Geuns, JMC. 1978. Steroid hormones and plant growth and development. Phytochemistry 17: 1-14.
Gibeaut, DM, Hulett, J, Cramer, GR, Seemann, JR. 1997. Maximal biomass of Arabidopsis thaliana using a simple, low-maintenance hydroponic method and favorable environmental conditions. Plant Physiol. 115: 317-319.
Hayat, S, Ahmad, A, Hussain, A, Mobin, M. 2001. Growth of wheat seedlings raised from the grains treated with 28-homobrassinolide. Acta Physiol. Plant. 23: 27-30.
Janeczko, A, Filek, W, Biesaga-Kościelniak, J, Marcińska, I, Janeczko, Z. 2003. The influence of animal sex hormones on the induction of flowering in Arabidopsis thaliana: comparison with the effect of 24-epibrassinolide. Plant Cell Organ Tissue Cult. 72: 147-151.
Orchard, BJ, Doucette, WJ, Chard, JK, Bugbee, B. 2000. Uptake of trichloroethylene by hybrid poplar trees grown hydroponically in flow-through plant growth chambers. Environ. Toxicol. Chem. 19: 895-903.
Ray, JD and Sinclair, TR. 1998. The effect of pot size on growth and transpiration of maize and soybean during water deficit stress. J. Exp. Bot. 49: 1381-1386.
115 Carmen G. Franks - Arabidopsis Phytoremediation
Shone, MGT and Wood, AV. 1974. A comparison of the uptake and translocation of some organic herbicides and a systemic fungicide by barley I. Absorption in relation to physico-chemical properties. J. Exp. Bot. 25: 390-400.
Thompson, PL, Ramer, LA, Schnoor, JL. 1998. Uptake and transformation of TNT by hybrid poplar trees. Environ. Sci. Technol. 32: 975-980.
Trapp, S, Zambrano, KC, Kusk, KO, Karlson, U. 2000. A phytotoxicity test using transpiration of willows. Arch. Environ. Contam. Toxicol. 39: 154-160.
116 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
CHAPTER 4 Method for detection of 17α-ethynylestradiol in surface water
4.1 INTRODUCTION
The synthetic hormone 17α-ethynylestradiol (EE2) is a known, potent endocrine
disrupting compound (EDC) and is a key contributor to the levels of estrogen and endocrine disruptors found within wastewater (Desbrow et al., 1998). Small concentrations of this hormone have been known to induce endocrine disruption effects in male trout (Routledge et al., 1998). Levels as low as 0.1 ng/ L have been reported to induce vitellogenin (precursor to yolk, a female-specific protein) production in male fish, as well as other sexual dysfunctions in fish (Purdom et al., 1994). Mammalian hormones that are also found within waste and surface waters are estrone (E1) and estradiol (E2).
Although not as potent as EE2, E1 and E2 are known EDC and are typically found in higher concentrations than EE2 in waste and surface waters (Desbrow et al., 1998).
Across Canada, mean concentrations of E2 and E1 in wastewater influent were 15.6 ng/ L and 49 ng/ L, and in final effluent they were reduced to 1.8 ng/ L and 17 ng/ L, respectively (Servos et al., 2005).
Reports of EE2 in the environment and wastewater have occurred around the world. A
nationwide survey within the United States detected a maximum level of 83.1 ng/ L and a
median level of 7.3 ng/ L of EE2 in streams (Kolpin et al., 2002). In Germany and the
U.K. levels of up to 17 ng/ L and 7 ng/ L, respectively, of EE2 have been reported in
wastewater effluent (Belfroid et al., 1999). Cargouët et al., (2004) reported EE2 levels
ranging from 1.0 to 3.2 ng/ L in wastewater treatment plant effluent and river surface
117 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
waters of the Paris area, France. Coastal surface waters of the German Baltic Sea were
examined for levels of estrogens, with levels of EE2 detected in a populated bay at 3.0
ng/ L in 2003 and 17.2 ng/ L in 2000 (Beck et al., 2005). Trace levels of EE2 have been
reported in the Waikato region of New Zealand (Sarmah et al., 2006). Not only is EE2
being detected in waste and surface waters, but also in drinking water. In 2001, Kuch and
Ballschmiter reported levels of up to 0.5 ng/ L of EE2 in tap water from a region of southern Germany that obtains its drinking water from a ground source.
Increasing concern over this compound, and other hormones and endocrine disrupting
compounds, has led to the development of methods of detection in wastewater influent
and effluent, as well as surface waters. Interest in the presence of this synthetic hormone
in local waters led to the development of a method for detection of EE2 specifically using
gas chromatography-mass spectrometry (GC-MS). GC-MS is a commonly used method
for the detection of pharmaceuticals and hormones in water and entails a relatively simple
stepped process of purification involving solid phase extraction (SPE) with C18 and silica
gel, and compound preparation involving derivatization, prior to GC-MS analysis (Kelly,
2000; Kuch and Ballschmiter, 2000; Mol et al., 2000; Jeannot et al., 2002; Hernando et
al., 2004; Shareef et al., 2004).
Silylation and GC-MS can result in the degradation of EE2 into E1 and E2 if performed improperly. This was not previously reported in the literature at the time of this method
development. Recovery levels of EE2 after improper silylation or GC-MS analysis has
the potential to result in low recovery levels of EE2. This degradation was noted during
118 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
this method development and a procedure was found that resulted in negligible
degradation. Improper preparation of EE2 is particularly a concern when a composite
detection method is employed, where more than one hormone is being detected in one
sample, leading to elevated levels of estrone and possibly undetectable levels of EE2.
This led to attempts at developing a method for the sole detection of EE2 to insure
minimal degradation. The method outlined here minimizes the degradation of EE2 into
E1 and optimizes the percentage of completely trimethylsilylated (TMS), di-TMS, versus
partially trimethylsilylated, mono-TMS, derivatives of EE2 for GC-MS analysis.
A method for detection and measurement of EE2 in river water was developed and
initially tested using river water samples and EE2 standard. Recoveries were low during
this procedure, possibly due to binding of EE2 to particulate matter resulting in losses
during clean-up steps. Improvement of this method was not concluded due to a
publication by Alberta Environment in late 2005 with preliminary analyses of the
Oldman River water for pharmaceuticals, including EE2, and other common wastewater
compounds (Sosiak and Hebben, 2005). This report found no detectable levels of EE2 within the Oldman River downstream of the City of Lethbridge wastewater treatment plant.
119 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.2 MATERIALS AND METHODS
4.2.1 Chemicals
Tritium-labeled ethynylestradiol, 17α-[6,7-3H(N)]-ethynylestradiol (3H-EE2) (specific activity: 40.0 Ci mmol-1; radiochemical purity >97% to HPLC analysis), was obtained
from PerkinElmer Life Sciences, Inc. (Boston, MA, USA). Unlabeled EE2 was obtained
from Sigma-Aldrich Canada Ltd. The chemical purity of EE2 was verified by nuclear
magnetic resonance (NMR) analysis (performed by Dr. Peter Dibble, Chemistry
Department, University of Lethbridge. See Appendix C for NMR results). The chemical
structure of EE2 is presented in Figure 4.1. Inclusion of 3H-EE2 was for tracing and
quantification purposes during the purification steps.
An internal standard was chosen that had a similar retention time (RT) and chemical
structure as EE2. The internal standard (IS), deuterated-17β-estradiol, (d2-E2) (98 atom
% D), was obtained from Sigma-Aldrich Canada Ltd. The chemical structure is presented in Figure 4.1. d2-E2 was chosen as an IS as levels of E2 within the surface water were not
known and may have been present. d2-E2 would also serve as a quantitative standard for
any E2 present.
The silylation reagent, BSTFA (N,O-bis(Trimethylsilyl)trifluoroacetamide) with 1%
TMCS (Trimethylchlorosilane) and the solvent pyridine were used for creation of trimethylsilyl (TMS) derivatives of the EE2 and d2-E2 (Pierce Biotechnology, Inc.,
Rockford, IL, USA).
120 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
All standard solutions were prepared in methanol due to poor solubility in water. Ethyl acetate was found to be a suitable solvent for transfer to GC-MS.
Solid phase extraction used octadecyl-functionalized silica gel (C18) (Sigma-Aldrich Co.)
and activated silica gel, 32-100 μm (Scientific Adsorbents Incorporated).
121 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
17α -Ethynylestradiol OH 17β -Estradiol-d2 OH CH3 CH3 C CH
H H D H H H H HO HO D 17-ethynyl-13-methyl-7,8,9,11,12,13,14,15,16,17 1,3,5[10]-estratriene-2,4-d -3,17β -diol 2 -decahydro-6H-cyclopenta[a]phenanthrene-3,17-diol
Si(CH33 ) OH* 17β -Estradiol-d2 O 17α -Ethynylestrad iol CH3 CH3 C CH
H H D H H H H O O (CH33 ) Si D (CH33 ) Si Partia lly trim ethylsilylated EE2 Comp letely trim ethylsilylated E2-d 2
Figure 4.1. Chemical structures of 17α-ethynylestradiol (EE2) and the IS 17β-estradiol- d2 (d2-E2) and their trimethylsilylated derivatives. The structure of partially trimethylsilylated EE2 is shown. The asterisk denotes other possible binding site for the
TMS group on EE2. For partial TMS derivatives, either site may be used. Completely trimethylsilylated EE2 would involve TMS binding to both of the hydroxyl oxygens.
Completely trimethylsilylated d2-E2 is shown.
122 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.2.2 Trimethylsilyl derivatives
Silylation derivatives, replacement of a hydroxyl hydrogen with a trimethylsilyl (TMS)
group (Si(CH3)3), were created in 3 mL reacti-vials. Addition of pyridine in a 1:1 volume ratio with the silylation reagent aided derivative formation. Figure 4.1 presents the
chemical structures of the TMS derivatives of EE2 and d2-E2.
A test standard of EE2 was prepared and dried down under nitrogen and low heat (35
°C). The vial was then removed from heat, 15 μL each of pyridine and
BSTFA+1%TMCS were added, the vial sealed, and heated to 60 °C for 30 minutes. The derivatives were removed from the heat and allowed to cool to room temperature for 2 hours prior to GC-MS analysis.
Initial GC-MS analysis was conducted with injection of the silylation reagents on-column
using the parameters described below. This method resulted in chromatograms containing
the derivatives of EE2 and also, unexpectedly, of E1. Figure 4.2 presents the chemical
structures of E1 and of TMS-E1 derivative. Discussion with Dr. P. Dibble suggested the
fluorine present in BSTFA is a reactive species capable of facilitating the breakdown of
EE2 into E1 during GC-MS analysis (personal communication with Dr. P. Dibble,
Chemistry Department, University of Lethbridge). Consequent removal of this reagent as the GC-MS injection solution assisted in negligible breakdown of EE2 during the GC run. The methods were modified so that the silylated sample was dried under nitrogen and low heat (35 °C) to remove the silylation reagents. Derivative samples were resuspended in 15 μL of ethyl acetate.
123 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Different combinations of temperature and time for creating complete TMS derivatives of
EE2 were attempted to determine the optimal conditions. Manufacturer’s suggested forcing conditions (150 °C for 12 hours) for complete silylation were also attempted at the early stages of method development, but these conditions were found to be extremely ineffective with degrading effects.
124 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Est ro n e O CH3
H
H H
HO 3-Hydroxy-1,3,5(10)-estratrien-17-one
Est ro n e O CH3
H
H H
O (CH33 ) Si Trim ethylsilyla ted E1
Figure 4.2. The chemical structure of estrone (E1) and of trimethylsilylated E1 is shown.
125 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.2.3 Gas chromatography-mass spectrometry
GC-MS determination was carried out on a HP5890A gas chromatograph coupled to a
HP5970B MSD in the electron ionization mode (EI). Derivative mixtures were injected cold ‘on-column.’
Column: DB5MS (J & W Scientific) 15 m; 0.25 mm i.d.; 0.25 um film thickness; head pressure 4 psi helium Precolumn: 50 cm; 0.53 um GC parameters: 60 °C – (0.50 min) - 20 °C/min – 200 °C – 5 °C/min – 300 °C – (5 min) MS parameters: transfer line at 280 °C; EI-mode (70 eV)
4.2.4 Calibration curve
For quantification of EE2 by GC-MS a calibration curve for EE2 and d2-E2 was created using single ion monitoring (SIM). Calibration standards ranged from 0.1, 0.2, 0.5, 1, 2,
5, and 10 ng for EE2 to 10 ng of IS. The EE2 standards and IS were dried down under nitrogen and low heat (35 °C). Silylation was carried out as described in section 4.2.2.
Standard mixtures were analyzed by GC-SIM. Characteristic ions and RTs are listed in
Table 4.1. Integrated area under the curve for SIM of EE2 and IS characteristic ions at m/z 425/418 were used to calculate ratios. The 3 to 4 other characteristic ions for each compound were monitored to confirm identity of the EE2 and d2-E2, and to check for the presence of E2 and E1.
126 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Table 4.1. SIM monitoring for EE2, E2, d2-E2, and E1 (TMS derivatives) characteristic ion m/z and retention times.
SIM m/z
EE2 E2 d2-E2 E1 RT (min) 15:90 14:20 14:57 14:30
Molecular ion 440 416 418 342
425 401 403 257
368 326 328 327
285 344 287 218
285 346 244 Note: EE2 and E2 both contain m/z 285
127 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.2.5 Preparative purification of Oldman River water
Sample collection and purification. The City of Lethbridge, Alberta, Canada, is located along the Oldman River. This river provides a source for drinking water and wastewater disposal, for not only Lethbridge, but other communities located along the river’s reach.
River water was collected during the month of July using sterilized containers. The containers were rinsed with river water prior to sample collection to neutralize active binding sites. Water was collected downstream of the drinking water intake diversion, but upstream of the wastewater outflows. The water was stored at 4 °C until its use in the experiments. As this preliminary method was not to detect levels of EE2 within the river water, but to prepare an analytical technique for the detection of hormones in surface waters, no preservative was added to the collected samples used to reduce the degree of microbial degradation (Labadie and Budzinski, 2005). The method outlined here was modified from previously reported methods (Kelly, 2000; Kuch and Ballschmiter, 2000;
Jeannot et al., 2002; Hernando et al., 2004; Petrovic et al., 2004).
Initial sample preparation. The water sample was allowed to warm to room temperature and suction filtered to remove large particulates using filter paper (Whatman No.1). Five litres were filtered and the pH adjusted to 8 using a 1% HCl solution.
Solid phase extraction (SPE). Two - 1 g C18 columns were prepared in 4 mL glass SPE tubes. The C18 was suspended in methanol, the column was packed and a thick, glass microfibre filter (GF/D) circle was gently placed on top of the column (to provide another
128 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
layer of filtration, as well as protection for the C18 column). The column was conditioned
with water.
A solution of EE2 was prepared (16,667 Bq 3H-EE2 + 10 ng EE2 in 10% aqueous
methanol as described in section 4.2.2) and divided between the two columns by addition to the top of the column and allowed to move into the C18 (gravity movement).
Figure 4.3 provides the schematic diagram of the solid phase extraction system used in
this research. The system was sealed and under suction. Two and a half litres were
allowed to run through each C18 column and collected.
The C18 columns were each then eluted with 3 mL 80 % aqueous methanol, the 2 eluants
pooled and checked again for radioactivity using a liquid scintillation counter (LSC) to
confirm recovery levels prior to the next purification step.
129 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
sample supply
column 1 column 2
collection container vacuum
Figure 4.3. Schematic diagram of the solid phase extraction system for surface waters.
130 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
A 1-g silica gel column was setup like the C18 columns. The silica was equilibrated with
hexane. The C18 eluant was dried under vacuum and low heat (35 °C). The sample was
resuspended in 25 μL ethyl acetate, then 225 μL of hexane was added to create a 90:10
solvent ratio. This was added to the top of the silica column, and the vial rinsed with
another 25 μL of ethyl acetate, to resuspend any residue, and the 90:10 solvent ratio created with 225 μL of hexane. Three solvent ratios were found to be sufficient to remove the compounds of interest from the silica column, with the analytes coming off primarily within the second solvent ratio (50:50% hexane : ethyl acetate):
90:10% hexane : ethyl acetate, 6 mL 50:50% hexane : ethyl acetate, 6 mL 100% ethyl acetate, 6 mL
The eluant was dried under partial vacuum and low heat (35 °C), the sample transferred
to a 3 mL reacti-vial. At this stage the d2-E2 was added to serve as a semi-quantitative IS for the subsequent GC-MS step. The combined eluant and IS were dried down under nitrogen and low heat in preparation for silylation (section 4.2.2).
GC-MS analysis. The procedure for GC-MS analysis of the river water samples is
described in section 4.2.3. A full scan and an SIM analysis (m/z listed in Table 4.1) were performed for the prepared river water samples.
The experiment was carried out twice. The IS d2-E2 was available only for the second
trial and was therefore not present in the first trial.
131 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.3 RESULTS
Trimethylsilyl derivatives and gas chromatography-mass spectrometry. Initial attempts
at GC-MS analysis of EE2 showed the presence of estrone (E1) and partial and complete
TMS-EE2 derivatives (Figure 4.4). The mass spectra of TMS-E1 and mono- and di-TMS-
EE2 are shown in Figure 4.4a and 4.4b.
Temperatures higher than 60 °C resulted in degradation of EE2 into E1 (Table 4.2).
Temperature and time of 60 °C for 30 minutes with 1:1 BSTFA+1%TMCS and pyridine
ratio and an injection solvent of ethyl acetate, resulted in a high percentage of complete
TMS derivatives for both EE2 and the IS, and low degree of degradation (Figure 4.5).
Table 4.2 presents some of the temperature, time and GC-MS injection solution
combinations tested for creating TMS derivatives of EE2. Percentages of total area of the
TIC peaks for the product of EE2 (E1) and TMS EE2 derivatives are presented.
132 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
15:90 TIC: 02.D
4000000
3800000
3600000
3400000
3200000
3000000
2800000
2600000
2400000 2200000 15:20 2000000
1800000
1600000
1400000 Relative Abundance 1200000
1000000
800000 600000 14:20 400000
200000
0 6.00 7.00 8.00 9.0010.0011.0012.0013.0014.0015.0016.0017.0018.0019.00 Time (minutes)
Figure 4.4. TIC of the products of EE2 produced during the silylation process and GC-
MS analysis. EE2 was silylated at 100 ° C for 60 minutes and allowed to sit for 2 hours prior to GC-MS analysis in ethyl acetate. RTs: mono-TMS-EE2, 15:20; di-TMS-EE2,
15:90; TMS-E1, spreading peak at 14:20-15:00.
133 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Table 4.2. Results from different temperature and time combinations for silylation of
EE2 presented as the percentage of total integrated area for TMS-E1 and mono-/di-TMS-
EE2 from the TICs.
% of Total E1 EE2 EE2 Combination GC-MS Injection TMS mono-TMS di-TMS
100 °C, 60 min Silylation solution 74 17 9
100 °C, 60 min Ethyl acetate 22 33 45
100 °C, 60 min Ethyl acetate 23 45 31
100 °C, 30 min Silylation solution 53 39 8
100 °C, 30 min Ethyl acetate 36 63 1
68 °C, 30 min Ethyl acetate 3 28 70
60 °C, 60 min Ethyl acetate 7 3 90
60 °C, 30 min Ethyl acetate 0 2 98
134 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Average of 14.144 to 15.094 min.: 05.D 240000 342 220000 mono-TMS-E1 200000 180000 160000 140000 120000 100000 257
Relative Abundance 80000 73 60000 218 40000 244 327 20000 55 115128 177192 231 285 91 149163 205 271 299314 0 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 m/z
Figure 4.4a. Mass spectrum of TMS-E1 at 14:20-15:00 minutes from Figure 4.4.
Molecular ion 342.
135 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Average of 15.816 to 16.021 min.: 03.D 320000 425 73 300000
280000 di-TMS-EE2 260000
240000
220000
200000
285 180000
160000
140000
120000
100000 Relative Abundance 196 232 80000
60000
40000 167 257 91 115 322 350 20000 55 141 215 442 302 397 369 0 60 80 100120140160180200220240260280300320340360380400420440 m/z
Average of 15.034 to 15.263 min.: 03.D 285 320000
300000
280000 mono-TMS-EE2 260000
240000 368 220000
200000
180000
160000
140000
120000 73
Relative Abundance 100000
80000
60000 232 300 342
40000 218 257
115 20000 55 91 177 203 271 129 163 149 327 0 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 m/z
Figure 4.4b. Mass spectra of mono- and di-TMS-EE2 from Figure 4.4. Molecular ion
440 for di-TMS-EE2 and 368 for mono-TMS-EE2.
136 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
15:90 TIC: 01.D 2e+07
1.9e+07
1.8e+07
1.7e+07
1.6e+07
1.5e+07
1.4e+07
1.3e+07
1.2e+07
1.1e+07
1e+07
9000000
8000000
7000000 Relative Abundance 6000000
5000000
4000000
3000000
2000000
1000000
0 6.00 7.00 8.00 9.0010.0011.0012.0013.0014.0015.0016.0017.0018.0019.00 Time (minutes)
Figure 4.5. TIC of di-TMS-EE2 derivative. EE2 TMS derivatives were prepared at 60°C for 30 minutes and GC-MS analysis was in ethyl acetate.
137 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Calibration curve. Ions for use in the calibration curve were selected from the full mass
spectra of TMS-EE2 (Figure 4.4b) and TMS-d2-E2 (IS) (Figure 4.7a). The calibration
curve for EE2 to d2-E2 was created using the area ratio of ions at m/z 425 to m/z 418,
respectively. An example of single ion chromatograms used for the area calculations is
shown in Figure 4.6. Figure 4.7 presents the TIC for 10 ng IS to 10 ng EE2. The mass
spectrum of di-TMS-d2-E2 is presented in Figure 4.7a. Figure 4.8 displays the area ratio
calibration curve for EE2 to IS using the characteristic masses 425/418, EE2 and d2-E2, respectively, with a linear regression R2 of 0.997.
In regards to the calibration of EE2 and the IS, initial thoughts were that EE2 to d2-E2 would be close to the same in relative abundance at the 1:1 ratio (10 ng/ 10 ng) in full scan analysis, but this was not the case. The TIC in Figure 4.7 shows the differences in peak heights for EE2 and d2-E2 at the 1:1 ratio. Consequently, standards were prepared a
second time and the calibration curve repeated with similar results.
There were no problems creating di-TMS-d2-E2 derivatives when using the optimal
temperature and time for di-TMS-EE2 derivatives determined in section 4.2.2.
138 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Ion 425.00 (424.70 to 425.70): 07.D 100000
80000
60000
40000 15:90
20000
0 14.00 14.20 14.40 14.60 14.80 15.00 15.20 15.40 15.60 15.80 16.00 16.20 16.40 me--> undance
14:59 Ion 418.00 (417.70 to 418.70): 07.D 100000 Relative Abundance
80000
60000
40000
20000
0 14.00 14.20 14.40 14.60 14.80 15.00 15.20 15.40 15.60 15.80 16.00 16.20 16.40 Time (minutes)
Figure 4.6. Single ion chromatograms from GC-SIM analysis of 10 ng di-TMS-EE2
(m/z 425, RT 15:90) to 10 ng di-TMS-d2-E2 (m/z 418, RT 14:59).
139 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
12:25 TIC: 01.D 550000
500000
450000
400000 14:59 350000 15:90 300000
250000
200000 Relative Abundance
150000
100000
50000
0 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 Time (minutes)
Figure 4.7. TIC from GC-MS (full scan) analysis of 10.0 ng di-TMS-EE2 (RT 15:90) and 10.0 ng IS (di-TMS-d2-E2) (RT 14:57). Note: Peak at ~ 12.25 min corresponds to the
mass spectrum of impurities found to be within the pyridine.
140 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Average of 14.464 to 14.561 min.: 01.D 418 14000
13000 di-TMS-d2-E2 12000
11000
10000
9000 73
8000 287
7000 207 6000
5000 Relative Abundance
4000
3000 55 129 233 2000 93 149 328 1000 179 110 255 309 355 0 40 60 80 100120140160180200220240260280300320340360380400420 m/z
Figure 4.7a. Mass spectrum for the IS di-TMS-d2-E2 from Figure 4.7. Molecular ion
418.
141 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
0.35
0.3
0.25 418) y = 0.3047x m/z R2 = 0.997 0.2 425/ IS 425/ m/z 0.15
0.1 Area Ratio (EE2
0.05
0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Weight Ratio (ng EE2/ ng IS)
Figure 4.8. GC-MS calibration curve for EE2 and the IS d2-E2 using characteristic
masses 425/418, respectively.
142 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Preparative purification of Oldman River water. Two experimental runs with spiked river water were performed.
The initial GC-MS run had no IS, but the characteristic ions for EE2 were present. Figure
4.9 presents the TIC of the first run with the extracted ion chromatograms for EE2 and IS also shown. Table 4.3 presents the recoveries for EE2 during the purification steps as quantified by the proportion of 3H-EE2 recovered at each stage. For the first run, final recovery of EE2 was estimated by the amount of 3H-EE2 present in the sample prior to
GC-MS analysis, as the IS was not present. The second run was more successful with the
IS notably present in the GC-MS analysis allowing for quantification (Figure 4.10).
After each clean-up step involving C18 and silica a fraction of the eluant was analyzed with LSC to check for radioactive losses during these steps, with the assumption that radiolabeled-EE2 would behave similarly to non-labeled EE2 and be lost in similar proportions following similar mechanisms. In both trials initial river water clean-up through C18 resulted in complete recovery of the added radioactivity once the columns were stripped (Table 4.3). The C18 eluant was dried and resuspended prior to silica gel purification. The first experimental run noted significant losses after the silica gel step, so attempts to track the losses during the second run involved checking the radioactivity prior to the silica stage after the C18 eluants had been dried and resuspended in a known volume. Losses were noted after this dry down step. Losses were again noted after the silica gel step.
143 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
The final recovery results acquired by GC-MS for the second experimental run were
quantified using the IS and the calibration curve (Figure 4.10). Final recovery of the 10
ng of EE2 added in the initial stages of the second preparatory run was 2 ng (Table 4.3).
Recovery from the first run was suggested to be 3 ng prior to GC-MS analysis by
quantity of radio-labeled EE2 present in the sample after the silica gel step. GC-MS analysis of the first trial showed the presence of the characteristic ions for EE2 (Figure
4.9).
144 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Table 4.3. Spiked river water recovery. Percentage recovery of EE2 after each river water purification step based on quantity of radio-labeled EE2 present and by the ratio of
IS to non-labeled EE2 from the GC-MS analysis.
% Recovery after % Recovery after % Recovery Final % Recovery drying & Run C after silica after 18 resuspending C clean-up 18 gel clean-up GC-MS fractions
1 100 N/A 34 N/A*
2 100 68 49 20
*EE2 was present in the GC-MS analysis, but the IS was not added.
145 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
TIC: 02.D (*)
8000 A 6000
4000
2000
0 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 > ance
Ion 440.00 (439.70 to 440.70): 02.D (*) 15:90 B 8000
6000
4000
2000
0
Relative Abundance 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 > ance
Ion 425.00 (424.70 to 425.70): 02.D (*)
8000 15:90 C
6000
4000
2000
0 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 > Time (minutes)
Figure 4.9. GC-MS (full scan) analysis for the first spiked river water trial. A, TIC; B,
C, extracted ion chromatograms for EE2 (m/z 440 and 425, RT 15:90) are shown. Note the relative abundances for each plot have been normalized so that the largest peak is shown at full scale in each case.
146 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
TIC: 01.D 3 200000
3000000
2800000
2600000 A
2400000
2200000
2000000
1800000
1600000
1400000 1200000 15:90
Relative Abundance 1000000
800000 14:57
600000
400000
200000
0 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 Time (minutes) B
Ion 440.00 (439.70 to 440.70): 01.D (*) 15:90 8000 6000 4000 2000 0 13.6013.8014.0014.2014.4014.6014.8015.0015.2015.4015.6015.8016.0016.20 -> dance C Ion 425.00 (424.70 to 425.70): 01.D (*) 15:90
8000 6000 4000 2000 0 13.6013.8014.0014.2014.4014.6014.8015.0015.2015.4015.6015.8016.0016.20
Relative Abundance -> dance 14:57 Ion 418.00 (417.70 to 418.70): 01.D (*) D 8000 6000 4000 2000 0 13.6013.8014.0014.2014.4014.6014.8015.0015.2015.4015.6015.8016.0016.20 Time (minutes)
Figure 4.10. GC-SIM analysis of second spiked river water trial, with IS evident. A,
TIC; B, C, D, single ion chromatograms for EE2 (m/z 440 and 425, RT 15:90) and the IS
(m/z 418, RT 14:57). Note the different time ranges for single ion chromatograms and the
TIC, as well relative abundances for each plot have been normalized so that the largest peak is shown at full scale in each case.
147 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
4.4 DISCUSSION
The methods outlined here are designed to selectively detect and measure the synthetic
hormone EE2 in surface waters. Previously reported methods tended towards detection of
a complex mixture of natural and synthetic estrogens, such as EE2, E1, and E2 (Kuch and
Ballschmiter, 2000; Jeannot et al., 2002; Kolpin et al., 2002; Petrovic et al., 2004;
Vethaak et al., 2005). This broad detection method likely resulted in previously
undocumented degradation of EE2 into E1, with the potential for an underestimation of
EE2 and overestimation of E1 in sampled waters.
A method for preparation of EE2 for GC-MS analysis minimizing degradation processes
was developed. This process resulted in near 100% completely trimethylsilylated (di-
TMS) EE2 derivatives with no breakdown into E1 (Figure 4.5). Creating the TMS
derivatives was optimized when using the solvent pyridine at a ratio of 1:1 with
BSTFA+1% TMCS. Heating at 60 °C for 30 minutes optimized complete TMS-EE2 (di-
TMS) derivatives (Figure 4.5 and Table 4.2). Removal of the silyl reagent BSTFA prior to the GC-MS run aided in reducing the breakdown of EE2, as fluorine present in the
BSTFA may be reactive enough to facilitate the breakdown (Dr. P. Dibble, personal
communication).
The purification process of river water for the quantification of the synthetic hormone
EE2 appears to be sufficient, as evidenced by the relatively clean TIC in Figure 4.9. SPE
involving two-1g C18 columns, with a volume of 2.5 litres of river water per column, did
not exceed the capacity of the column (and therefore not allowing EE2 to pass through
148 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
the column) as suggested by 100% recovery of the EE2 test standards off the C18 columns
(Table 4.3). Final recovery was less than 70% of the added 10 ng/ L EE2 (Table 4.3 and
Figure 4.10).
There is some evidence that EE2 adsorbs to particulate matter during purification with
SPE and dry-down steps. This has been reported by Xiao et al. (2001) suggesting
estrogen has capabilities of adsorbing to suspended particulate matter that may be
removed during SPE, but not desorbed simply by rinsing with organic solvents. Lai et al.
(2000) found synthetic estrogens to sorb to sediments and suspended particulates at a
greater percentage than natural estrogens, with a general dependence on organic carbon
content and particle size. Losses of radio-labeled EE2 tracer presented in Table 4.3 are
consistent with these findings.
A method published by Ingrand et al. (2003) using LC-MS-MS ion trap system also noted losses during each preparation step for spiking mineral water, including SPE on
C18, liquid-liquid separation, and a clean-up step, ending with a total recovery of 87%.
When using spiked wastewater effluent EE2 losses were substantial, resulting in levels lower than limits of quantification (10 ng/ L) when spiked with 10 or 20 ng/ L.
EE2 is typically reported at levels lower than detection, presumably due to the higher lowest level of detection and the smaller concentration being released into the wastewater. Natural estrogens are commonly found, with E1 typically at higher quantities than other natural estrogens. Also, degradation of these hormones during wastewater
149 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
treatment could result in E1 and the resulting higher quantities being detected. E2 is known to degrade into E1 by using nitrifying activated sludge (Ternes et al., 1999a and
1999b; Shi et al., 2004).
In 2004, Shareef et al. also noted the degradation of EE2 into E1 during certain
experimental procedures. In the fall of 2005, Zhang and Zuo published an article on the
degradation of EE2 into E1 during the process of creating derivatives using BSTFA +
TMCS, as well as the problem of incomplete TMS-EE2 derivatives. Suggested methods
for optimizing creation of TMS derivatives of EE2 are the same as those determined by
the methods presented here.
In the time since this method was started, an improved method for the detection of EE2 in
surface waters was published. Noppe et al., 2005, developed a method with GC-EI-MS-
MS with full recoveries of standards. Similar to other methods, the initial purification of
the water was performed with SPE extraction. Noppe et al. (2005) preferred the use of
discs (using Bakerbond Speedisk octadecyl-bonded silica (C18XF)) over cartridges to
limit the amount of clogging from colloidal material and suspended particles in the
environmental samples. Clogging was found to not be a problem for the SPE methods
used here. The pH of their water samples were adjusted to 7 to minimize ionized organic
acids, although they found a pH range of 2 - 7 was optimal for recovery. The methods
developed here used an adjusted pH of 8 to ensure the analyte and organic acids were
non-ionized. It may be that a neutral pH was needed to minimize adsorption of EE2 to
particulate matter.
150 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
At the time this method was being developed, Alberta Environment published a preliminary report examining pharmaceuticals and wastewater contaminants in the province of Alberta, including EE2 (Sosiak and Hebben, 2005). The Oldman River, downstream of the City of Lethbridge’s wastewater treatment plant, was tested. Levels of
EE2 were below levels of detection which ranged from 0.022 to 0.094 ng/ L across standards and between laboratories performing the analyses. The publication checked major Alberta cities wastewater effluent for 105 compounds and isomer mixtures. EE2 was not detected at other major cities, except for the City of Calgary with one wastewater treatment plant’s effluent at a level of 8 ng/ L.
151 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
LITERATURE CITED
Beck, I-C, Bruhn, R, Gandrass, J, Ruck, W. 2005. Liquid chromatography-tandem mass spectrometry analysis of estrogenic compounds in coastal surface water of the Baltic Sea. J. Chromatogr. A 1090: 98-106.
Belfroid, AC, Van der Horst, A, Vethaak, AD, Schäfer, AJ, Rijs, GBJ, Wegener, J, Cofino, WP. 1999. Analysis and occurrence of estrogenic hormones and their glucuronides in surface water and waste water in The Netherlands. Sci. Total Environ. 225: 101-108.
Cargouët, M, Perdiz, D, Mouatassim-Souali, A, Tamisier-Karolak, S, Levi, Y. 2004. Assessment of river contamination by estrogenic compounds in Paris area (France). Sci. Total Environ. 324: 55-66.
Desbrow, C., Routledge, EJ, Brighty, G, Sumpter, JP, Waldock, MJ. 1998. Identification of Estrogenic Chemicals in STW Effluent. 1. Chemical Fractionation and in Vitro Biological Screening. Environ. Sci. Technol. 32: 1546-1558.
Hernando, MD, Mezcua, M, Gómez, MJ, Malato, O, Agüera, A, Fernándex-Alba, AR. 2004. Comparative study of analytical methods involving gas chromatography-mass spectrometry after derivatization and gas chromatography-tandem mass spectrometry for the determination of selected endocrine disrupting compounds in wastewaters. J. Chromatogr. A 1047: 129-135.
Ingrand, V, Herry, G, Beausse, J, de Roubin, M-R. 2003. Analysis of steroid hormones in effluents of wastewater treatment plants by liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 1020: 99-104.
Jeannot, R, Sabik, H, Sauvard, E, Dagnac, T, Dohrendorf, K. 2002. Determination of endocrine-disrupting compounds in environmental samples using gas and liquid chromatography with mass spectrometry. J. Chromatogr. A 974: 143-159.
Kelly, C. 2000. Analysis of steroids in environmental water samples using solid-phase extraction and ion-trap gas chromatography-mass spectrometry and gas chromatography-tandem mass spectrometry. J. Chromatogr. A 872: 309-314.
Kolpin, DW, Furlong, ET, Meyer, MT, Thurman, EM, Zaugg, SD, Barber, LB, Buxton, HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci. Technol. 36: 1202-1211.
Kuch, HM and Ballschmiter, K. 2000. Determination of endogenous and exogenous estrogens in effluents from sewage treatment plants at the ng/L-level. Fresenius J. Anal. Chem. 366: 392-395.
152 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
Kuch, HM and Ballschmiter, K. 2001. Determination of endocrine-disrupting phenolic compounds and estrogens in surface and drinking water by HRGC-(NCI)-MS in the picogram per liter range. Envion. Sci. Technol. 35: 3201-3206.
Labadie, P and Budzinski, H. 2005. Development of an analytical procedure for determination of selected estrogens and progestagens in water samples. Anal. Bioanal. Chem. 381: 1199-1205.
Lai, KM, Johnson, KL, Scrimshaw, MD, Lester, JN. 2000. Binding of waterborne steroid estrogens to solid phases in river and estuarine systems. Environ. Sci. Technol. 34: 3890-3894.
Mol, HGJ, Sunarto, S, Steijger, M. 2000. Determination of endocrine disruptors in water after derivatization with N-methyl-N-(tert.-butyldimethyltrifluoroacetamide) using gas chromatography with mass spectrometric detection. J. Chromatogr. A 879: 79- 112.
Noppe, H, De Wasch, K, Poelmans, S, Van Hoof, N, Verslycke, T, Janssen, CR, De Brabander, HF. 2005. Development and validation of an analytical method for detection of estrogens in water. Anal. Bioanal. Chem. 382: 91-98.
Petrovic, M, Eljarrat, E, Lopez de Alda, MJ, Barceló, D. 2004. Endocrine disrupting compounds and other emerging contaminants in the environment: A survey on new monitoring strategies and occurrence data. Anal. Bioanal. Chem. 378: 549-562.
Purdom, CE, Hardiman, PA, Bye, VJ, Eno, NC, Tyler, CR, Sumpter, JP. 1994. Estrogenic effects of effluents from sewage treatment works. Chem. Ecol. 8: 275- 285.
Routledge, EJ, Sheahan, D, Desbrow, C, Brighty, GC, Waldock, M, Sumpter, JP. 1998. Identification of Estrogenic Chemicals in STW Effluent. 2. In vivo responses in trout and roach. Environ. Sci. Technol. 32: 1559-1565.
Sarmah, AK, Northcott, GL, Leusch, FDL, Tremblay, LA. 2006. A survey of endocrine disrupting chemicals (EDCs) in municipal sewage and animal waste effluents in the Waikato region of New Zealand. Sci. Total Environ. 355: 135-144.
Servos, MR, Bennie, DT, Burnison, BK, Jurkovic, A, McInnis, R, Neheli, T, Schnell, A, Seto, P, Smyth, SA, Ternes, TA. 2005. Distribution of estrogens, 17β-estradiol and estrone, in Canadian municipal wastewater treatment plants. Sci. Total Environ. 336: 155-170.
Shareef, AC, Parnis, CJ, Angove, MJ, Wells, JD, Johnson, BB. 2004. Suitability of N,O- bis(trimethylsilyl)trifluoroacetamide and N-(tert-butyldimethylsilyl)-N- methyltrifluoroacetamide as derivatization reagents for the determination of the
153 Carmen G. Franks - 17α-Ethynylestradiol Detection Method
estrogens estrone and 17alpha-ethinylestradiol by gas chromatography-mass spectrometry. J. Chromatogr. A 1026: 295-300.
Shi, J, Fujisawa, S, Nakai, S, Hosomi, M. 2004. Biodegradation of natural and synthetic estrogens by nitrifying activated sludge and ammonia-oxidizing bacterium Nitrosomonas europaea. Water Res. 34: 2323-2330.
Sosiak, A and Hebben, T. 2005. A preliminary survey of pharmaceuticals and endocrine disrupting compounds in treated municipal wastewaters and receiving waters of Alberta. Environmental Monitoring and Evaluation Branch. Alberta Environment. Pub. No: T/773. 64 p. Available online: http://www3.gov.ab.ca/env/info/infocentre/publist.cfm
Ternes, TA, Kreckel, P, Mueller, J. 1999. Behaviour and occurrence of estrogens in municipal sewage treatment plants - II. Aerobic batch experiments with activated sludge. Sci. Total Environ. 225: 91-99.
Vethaak, AD, Lahr, J, Schrap, SM, Belfroid, AG, Rijs, GBJ, Gerritsen, A, de Boer, J, Bulder, AS, Frinwis, GCM, Kuiper, RV, Legler, J, Murk, TAJ, Peijnenburg, W, Verhaar, HJM, de Voogt, P. 2005. An integrated assessment of estrogenic contamination and biological effects in the aquatic environment of The Netherlands. Chemosphere 59: 511-524.
Xiao, X-Y, McCalley, DV, McEvoy, J. 2001. Analysis of estrogens in river water and effluents using solid-phase extraction and gas chromatography-negative chemical ionization mass spectrometry of the pentafluorobenzoyl derivatives. J. Chromatogr. A 923: 195-204.
Zhang, K and Zuo, Y. 2005. Pitfalls and solution for simultaneous determination of estrone and 17α-ethynylestradiol by gas chromatography-mass spectrometry after derivatization with N,O-bis(trimethylsilyl)trifluoroacetamide. Anal. Chim. Acta 554: 190-196.
154 Carmen G. Franks - Discussion
CHAPTER 5 Discussion
5.1 PHYTOREMEDIATION OF PHARMACEUTICALS AND HERBICIDE
Pharmaceuticals are designed to be chemically stable to increase shelf life, and provide
persistence in the metabolic environments. Consequently, it then should be no surprise
that pharmaceuticals have been found, in both parental and metabolite form, in urine and
feces making their way through wastewater treatment facilities to be disposed of often in
surface waters. This deliberate drug stability therefore increases the prospective impact on non-target organisms and their ecosystems.
Natural, aesthetic, and cost efficient, phytoremediation with aquatic or phreatophytic
plants may provide a method for removing pharmaceuticals from wastewater and the
water environment. Using the log of the octanol-water coefficient (log Kow) of the
pharmaceuticals, estimates can be made into whether the compounds will be taken up by plants, how much could be taken up, and where they will be distributed within the plants.
Each plant species can vary in the amount of water uptake and phytoremediation
capabilities. Also, each pharmaceutical will have a different log Kow and be affected
differently by the environment, plants and chemical interactions with other compounds.
Compounds with log Kow values generally above and below 2 will be less likely to be taken into and transported within the plants.
Members of the Salicaceae family, particularly of the genus Populus commonly referred
to as poplar trees, have been the plant of choice for most phytoremediation research and
155 Carmen G. Franks - Discussion
field applications. Offering ample geographic distribution, fast growth, deep roots and
extensive transpiration, these trees have been the focus in several areas of
phytoremediation. Research into their application has been reported with oils and
hydrocarbons (Widdowson et al., 2005), herbicides (Burken and Schnoor, 1996 and
1997), pesticides (Karthikeyan et al., 2004) and explosives (Thompson et al., 1998), synthetic compounds (Kassel et al., 2002), and various organic pollutants (Burken and
Schnoor, 1998).
This plant family also includes the genus Salix, the willows. Although often ignored in
the field of phytoremediation, this genus shares physiological characteristics of Populus.
These shared traits, and family, would suggest willow to behave similarly to poplars in their abilities to remediate. Further insight into the genes and metabolic processes involved in Populus phytoremediation can now be investigated as a model for the processes within willow since the Populus genome has been sequenced (Sterky at al.,
2004). The willows are just now beginning to be viewed as potential candidates for phytoremediation. Recent studies involving uptake of contaminants and willow include cyanide (Larsen et al., 2004; Larsen and Trapp, 2006), the ‘antifoulant’ tributyltin
(Ciucani et al., 2004), cadmium (Lewandowski et al., 2006), as well as increased levels of polychlorinated biphenyl-degrading bacteria in the root soil of willows (Leigh et al.,
2006). These studies revealed willow to be a promising remediation plant, especially for water contaminants since willows are exceptionally inundation tolerant and typically the woody plants closest to the streams in riparian zones.
156 Carmen G. Franks - Discussion
Arabidopsis thaliana is a plant that would not normally be considered for a remediation
project. A member of the Brassica family, Arabidopsis has been extensively studied and
its genome sequenced. Determining the physiological genetics of uptake, transport and
metabolism enzymes involved in detoxification of pharmaceuticals using Arabidopsis can be an important step towards better understanding phytoremediation, the processes at work and which plants might be more suited to certain remediation projects. For example, Arabidopsis has been used in transgenic studies with the insertion of a bacterial enzyme gene, mercuric ion reductase, improving the plants tolerance to mercuric ions and its abilities to convert it into less toxic elemental mercury (Dietz and Schnoor, 2001).
This MSc project set out to investigate the abilities of Salix exigua and Arabidopsis
thaliana at uptake of three common pharmaceuticals from solution, followed by subsequent analyses examining the transport and apparent forms with which the pharmaceuticals were distributed within the plants. Both plants were effective at removing and transporting these compounds from solution. Of the compounds taken into the plants, bound and soluble forms were evident. An empirical relationship between compound octanol-water partition coefficient and uptake, distribution and the plant species was able to be developed. The determined empirical relationships were found to be similar to reported empirical relationships for barley and poplar and supported the theory that predictable behaviours for neutral pharmaceuticals can be based on their octanol-water partition coefficient.
157 Carmen G. Franks - Discussion
Although there are major ecophysiological differences, it was interesting to note some of
the similarities between willow and Arabidopsis. Particularly because some of the parameters measured (uptake, RCF, TSCF) depend on root and shoot masses of the
experimental plants, a valid comparison of these parameters requires that root and shoot
masses be similar. Arabidopsis shoot weights were significantly larger than for willow
across compounds (Table 5.1 and Table A5.1, ANOVA, p < 0.05). Root fresh weight was
similar for Arabidopsis and willow except for DTZ in which willow root weight was less
than for Arabidopsis (p = 0.048). These similarities in fresh root weight allow for some
comparisons among Arabidopsis and willow.
In comparing final cumulative transpiration between the plants and for each compound
(Table 5.2 and Table A5.2, ANOVA) there were no significant differences. Even though
shoot mass for Arabidopsis was significantly greater than for willow for all of the
compounds, the final volumes transpired did not vary.
In comparing the percentage uptake over time between the plants and for each compound
(Table A5.3a-d, ANOVA) a few differences occur. For DTZ, Arabidopsis uptake was
significantly lower than with willow at each sampling time (ANOVA, p < 0.05). DZP
uptake between plants was not significantly different except at t = 8 with Arabidopsis
uptake significantly lower than with willow (ANOVA, p = 0.036). Uptake of ATZ varied
between plants only at t = 2 with willow uptake significantly lower than with Arabidopsis
(ANOVA, p < 0.05). Uptake over time barely differs among the plants and three of the
158 Carmen G. Franks - Discussion
compounds, excluding DTZ, which is likely a function of root mass for the relatively
short 24 hour period of time these studies were carried out.
The final percentage uptake, or removal from solution after the 24 hour study period
between Arabidopsis and willow were similar except for DTZ, which was significantly lower for Arabidopsis (Table 5.2 and Table A5.1, ANOVA, p < 0.05). These similar final uptake values are likely a function of the similar root masses among plants (Table A5.1).
The difference noted in DTZ is difficult to explain. It is unclear why DTZ would apparently behave as a neutral compound with one plant and an ionized compound with the other.
Uptake from solution did follow the predicted order based on the compounds log Kow
values. EE2, having the highest log Kow of the four compounds was removed at a higher
percent than the other compounds. Next was DTZ, DZP, then ATZ, with corresponding
log Kow of 2.70, 2.82, and 2.61. If the compounds were passive, and not ionized, then one
might assume removal would correspond to the Kow. DTZ, though, can be ionized at
environmental and physiological pH and would become hydrophobic in the apoplast and
pass the membrane easily (likely explaining its higher percentage of uptake over the
higher Kow DZP within willow).
159 Carmen G. Franks - Discussion
Table 5.1. Salix exigua and Arabidopsis root and shoot fresh weight and total volume transpired over the 24 hour study period for three pharmaceuticals and the herbicide
(mean ± SE). EE2, 17α-ethynylestradiol; DTZ, diltiazem; DZP, diazepam; and ATZ, atrazine.
Salix exigua Arabidopsis thaliana
Root Shoot Total Vol. Root Shoot Total Vol. fr. wt. (g) fr. wt. (g) Transp. (mL) fr. wt. (g) fr. wt. (g) Transp. (mL) 0.47 ± 0.05 0.71 ± 0.05 10.42 ± 0.58 EE2 0.79 ± 0.12 1.85 ± 0.25 9.65 ± 0.81 0.53 ± 0.10 0.93 ± 0.16 13.67 ± 2.39 DZP 0.70 ± 0.10 1.47 ± 0.21 8.83 ± 0.90 0.41 ± 0.13 0.64 ± 0.08 9.25 ± 0.46 DTZ 0.73 ± 0.10 1.65 ± 0.21 9.13 ± 0.97 0.48 ± 0.08 1.00 ± 0.09 14.25 ± 1.23 ATZ 0.67 ± 0.18 1.75 ± 0.17 10.85 ± 1.40
160 Carmen G. Franks - Discussion
Table 5.2. Percentage final uptake of three pharmaceuticals and an herbicide (0.04 μg/ mL) by Salix exigua and Arabidopsis after 24 hours (mean ± SE), with listed log Kow values.
Uptake (%)
Salix Arabidopsis Log K ow exigua thaliana
17a-Ethynylestradiol 3.67 87.8 ± 1.12 84.5 ± 0.01 Diazepam 2.82 62.5 ± 3.89 58.8 ± 0.03 Diltiazem 2.70 77.3 ± 6.12 56.9 ± 0.03 Atrazine 2.61 50.1 ± 3.24 52.0 ± 0.05
161 Carmen G. Franks - Discussion
Salix exigua and Arabidopsis plants underwent the same separation into components and
extraction procedures to determine distribution of the compounds within the plants.
Recovery levels were acceptable for willow, although ATZ recovery was poor, while
Arabidopsis recovery levels were significantly lower than for willow, except for EE2.
The reasons for the low and diverse recovery values are not understood, especially since a high degree of consistency for the proportions recovered for each replicate are evident while recoveries were more variable. There were procedural variations associated with oxidation and a possible explanation for the consistency in distribution among replicates may be because samples are oxidized in sequence (i.e. DZP replicate plant A root and shoot would be oxidized before replicate B). The same losses may have occurred on the entire replicate resulting in more accurate proportions.
Although Arabidopsis recoveries were low, replicate plant recoveries were similar
enough in proportion distribution to provide an idea of general distribution within the
plant. Sufficient consistency was achieved to enable comparison between Arabidopsis
and willow.
Distribution between roots and shoots among Arabidopsis and willow were comparable
between EE2, DZP and ATZ (Figures 2.6 and 2.7 and Figures 3.4 and 3.5). There were
also similarities in the proportions of bound and soluble fractions between the plants. For
EE2, most of the compound was found within the roots in a bound form in both willow
and Arabidopsis. DZP and ATZ were both transported to the shoots with small
162 Carmen G. Franks - Discussion
proportions becoming bound in roots and shoots. This again suggests that these
pharmaceuticals behave as predicted by their log Kow values. As the percentage
distributions between the compounds were similar (Figures 2.6 and 2.7 and Figures 3.4
and 3.5), the relationship between transpiration and uptake may exist and occur as
expected for neutral compounds. This correlation may be explained as TSCF. A correlation between uptake and cumulative transpiration trends for the individual plants was also not observed, except for Arabidopsis and EE2, likely due to the larger sample size and greater range of plant sizes.
DTZ distribution and uptake were significantly different between the plants. In
Arabidopsis, DTZ appeared to behave as predicted by its log Kow value (that of a neutral compound), with > 50% remaining within the root as soluble and bound fractions. In willow, DTZ apparently behaved more as an ionized compound and remained entirely soluble within the root.
Calculated RCF values for Arabidopsis were similar to those experimentally determined
for willow (Table 5.1). EE2 RCF values were not significantly different, whereas DTZ
and DZP were. Arabidopsis RCF to log Kow values fit a linear regression equation at a
similar, but more steeply increasing slope to willow (Table 5.3 and Figures 2.9 and 2.10).
This consistent trend suggests that the uptake of relatively neutral pharmaceuticals is
guided by their physiochemical properties, particularly their log Kow. Following similar
trends reinforces the idea of log Kow being important in predictions of chemical behaviour in the environment. Uptake or equilibrium curves from the 24 hour uptake studies and the
163 Carmen G. Franks - Discussion
RCF studies follow a similar curve, verifying that similar processes are taking place and
that RCF plays a role in whole plant uptake values (Figures 2.4 and 2.8 and Figure 3.2).
Empirical relationships between TSCF and log Kow for Arabidopsis and willow were very
similar in their downward trend with increasing log Kow values (Table 5.4 and Figures
2.11 and 3.11). The determined relationships for willow and Arabidopsis decline at a faster rate with increasing log Kow than relationships determined for barley and poplar
(Table 5.4). TSCF values for DZP and EE2 are equivalent between willow and
Arabidopsis, with ATZ and DTZ values being significantly higher for Arabidopsis than
willow. Although, the estimated TSCF value for willow and DTZ may be significantly
different due to the observed difference in behaviour of DTZ among the study plants.
If a TSCF of 1 is equivalent to passive uptake following the transpiration stream, then the
values determined for DZP, DTZ and ATZ with Arabidopsis suggest these compounds
move almost passively with water. The TSCF value for DZP with willow is also close to
the value of one. These values seem high, particularly when their log Kow values, when
compared with other compounds, would suggest a more restricted transport.
Comparing the RCF and TSCF values for willow and Arabidopsis to their uptake and distribution, the proportions of distribution of EE2, DZP and ATZ are almost identical
(Figures 3.4 and 3.5 and Figures 2.6 and 2.7). Considering the RCF values of willow for
ATZ and DZP were larger than for Arabidopsis (Table 5.3), it could be expected that a larger portion of the compound would be found within the roots of willow than in
164 Carmen G. Franks - Discussion
Arabidopsis. This is not the case. The proportions of recovered radioactivity within the roots and shoots for both ATZ and DZP are equivalent between the two plants.
165 Carmen G. Franks - Discussion
Table 5.3. Calculated and experimentally determined root concentration factor values
(mean ± SE) for Salix exigua and Arabidopsis. Also shown are the root concentration factor values obtained for the four compounds using their log Kow values and Briggs et al.
(1982) equation for barley, and Burken and Schnoor’s (1998) equation for hybrid poplar.
Root concentration factor
Log Kow S. exigua Arabidopsis barley poplar 17a-Ethynylestradiol 3.67 210 ± 50 219 ± 44 21.05 9.54 Diazepam 2.82 34 ± 7 8.9 ± 0.9 5.30 4.83 Diltiazem 2.70 257 ± 24 21 ± 5 4.44 4.53 Atrazine 2.61 22 ± 5 2.5 ± 0.57 3.91 4.34
166 Carmen G. Franks - Discussion
Table 5.4. Calculated and experimentally determined transpiration stream concentration values (mean ± SE) for Salix exigua and Arabidopsis. Also shown are the transpiration
stream concentration factor values obtained for the four compounds using their log Kow values and Briggs et al. (1982) equation for barley, and Burken and Schnoor’s (1998) equation for hybrid poplar.
Transpiration stream concentration factor
Log Kow S. exigua Arabidopsis barley poplar 17α-Ethynylestradiol 3.67 0.15 ± 0.06 0.20 ± 0.03 0.18 0.44 Diazepam 2.82 0.93 ± 0.15 0.93 ± 0.22 0.50 0.73 Diltiazem 2.70 0.00 ± 0.00 0.93 ± 0.20 0.55 0.74 Atrazine 2.61 0.73 ± 0.19 0.98 ± 0.20 0.59 0.75
167 Carmen G. Franks - Discussion
Summary discussion. This study suggests that phytoremediation could be an effective
option for removing pharmaceuticals within water environments. As both Salix exigua and Arabidopsis removed between 50 and 90% of the compounds from solution, it is probable that other plants are also capable of phytoremediation of pharmaceuticals. Both plants affirmed that uptake and transport of pharmaceuticals can be predicted based on their chemical properties, particularly their octanol-water partitioning coefficient.
Although willow and Arabidopsis both performed well within this experimental setup, a
hydroponic system can be far from what is actually found within the environment. Field
applications, such as monitoring wastewater treatment wetland vegetation for the
presence of these compounds, need to be performed to further confirm the usefulness of
willow in the field environment.
Further examination into the fate of these compounds once they have entered plants
needs to be considered to determine the metabolites and their potency and possible release rates upon ingestion or decay. This process of determining if metabolism occurs, and potential metabolites, could involve high performance liquid chromatography
(HPLC) analysis of the solvent-extracted soluble fractions followed by gas chromatography-mass spectrometry (GC-MS) analysis (the methods outlined in Chapter
4 could provide a framework for this process, particularly for EE2). Fate of these compounds within plants over time could be analyzed with time course experiments and longer periods of uptake prior to analysis (much as the preliminary investigation for Salix
168 Carmen G. Franks - Discussion
exigua and EE2 in Appendix B was performed). The effect of different concentrations of
these pharmaceuticals could be determined as well as the toxic levels.
Differences in environmental temperature, type of plants, age of plants and synergistic
pharmaceutical effects could also be investigated. Not only does further research into
phytoremediation of pharmaceuticals exist, but also into the methodologies for working with the pharmaceuticals. As was found in these experiments, the behaviour of pharmaceuticals with SPE, preparative steps prior to GC-MS analysis and GC-MS analysis, and losses incurred during dry-down steps during basic laboratory procedures suggests that methods of working with specific pharmaceutical needs to be determined prior to field inventories and experiments.
I thus conclude that willow is very effective at removing pharmaceuticals from solution
and could provide sequestration of these compounds in a bound form, limiting their re-
entry into the environment. Since willow is a member of the Salicaceae family and
performed similarly to prior research with poplar, poplar research may be applicable to
willow and vice versa. This study also confirms the predictable behaviour of
pharmaceuticals in the environment and plants based on their octanol-water partitioning
coefficient. The support of Arabidopsis confirming these predictions only affirms the
potential of plants for phytoremediation of the water environment.
169 Carmen G. Franks - Discussion
LITERATURE CITED
Briggs, GG, Bromilow, RH, Evans, AA. 1982. Relationships between lipophilicity and root uptake and translocation of non-ionised chemicals by barley. Pestic. Sci. 13: 495-504.
Burken, JG and Schnoor, JL. 1996. Phytoremediation: Plant uptake of atrazine and role of root exudates. J. Environ. Eng. 122: 958-963.
Burken, JG and Schnoor, JL. 1997. Uptake and metabolism of atrazine by poplar trees. Environ. Sci. Technol. 31: 1399-1406.
Burken, JG, and Schnoor, JL. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32: 3379-3385.
Ciucani, G, Mosbaek, H, Trapp, S. 2004. Uptake of tributyltin into willow trees. Environ. Sci. Pollut. Res. Int. 11: 267-272.
Dietz, AC and Schnoor, JL. 2001. Advances in phytoremediation. Environ. Health Perspect. 109: 163-168.
Karthikeyan, R, Davis, LC, Erickson, LE, Kassim Al-Khatib, LE, Kulakow, PA, Barnes, PL, Hutchinson, SL, Nurzhanova, AA.2004. Potential for plant-based remediation of pesticide-contaminated soil and water using nontarget plants such as trees, shrubs, and grasses. Crit. Rev. Plant Sci. 23: 91-101.
Kassel, AG, Ghoshal, D, Goyal, A. 2002. Phytorememdation of trichloroethylene using hybrid poplar. Physiol. Mol. Biol. Plants 8: 3-10.
Larsen, M and Trapp, S. 2006. Uptake of iron cyanide complexes into willow trees. Environ. Sci. Technol. 40: 1956-1961.
Larsen, Morten, Trapp, Stefan, Pirandello, Alessandro. 2004. Removal of cyanide by woody plants. Chemosphere 54: 325-333.
Leigh, MB, Prouzova, P, Mackova, M, Macek, T, Nagle, DP, Fletcher, JS. 2006. Polychlorinated biphenyl (PCB)-degrading bacteria associated with trees in a PCB- contaminated site. Appl. Environ. Microbiol. 72: 2331-2342.
Lewandowski, I, Schmidt, U, Londo, M, Faaij, A. 2006. The economic value of the phytoremediation function – assessed by the example of cadmium remediation by willow (Salix spp). Agric. Syst. 89: 68-89.
170 Carmen G. Franks - Discussion
Sterky, F, Bhalerao, RR, Unneberg, P, Segerman, B, Nilsson, P, Brunner, AM, Charbonnel-Campaa, L, Lindvall, JJ, Tandre, K, Strauss, SH, Sundberg, B, Gustafsson, P, Uhlen, M, Bhalerao, RP, Nilsson, O, Sandberg, G, Karlsson, J, Lundeberg, J, Jansson, S. 2004. A Populus EST resource for plant functional genomics. Proc. Natl. Acad. Sci. U. S. A. 101: 13951-13956.
Thompson, PL, Ramer, LA, Schnoor, JL. 1998. Uptake and transformation of TNT by hybrid poplar trees. Environ. Sci. Technol. 32: 975-980.
Widdowson, MA, Shearer, S, Andersen, RG, Novak, JT. 2005. Remediation of polycyclic aromatic hydrocarbons in groundwater using poplar trees. Eviron. Sci. Technol. 39: 1598-1605.
171 Carmen G. Franks - Appendices
APPENDIX A Statistical analyses
172 Carmen G. Franks - Appendices
Chapter 2
Table A2.1. Salix exigua ANOVA comparison among compounds for mean total volume transpired, root, shoot, wood and total fresh weight of replicate plants.
Sum of Mean df F Sig. Squares Square Between Groups .037 3 .012 .298 .826 Root fr. wt. (g) * (Combined) Compound Within Groups .747 18 .042 Total .784 21 Between Groups 1.111 3 .370 6.330 .004 Wood fr. wt. (g) * (Combined) Compound Within Groups 1.053 18 .058 Total 2.163 21 Between Groups .470 3 .157 2.479 .094 (Combined) Shoot fr. wt. (g) * Within Groups 1.139 18 .063 Compound Total 1.609 21 Between Groups 2.827 3 .942 2.961 .060 Total fr. wt. (g) * (Combined) Compound Within Groups 5.729 18 .318 Total 8.556 21 Between Groups 91.924 3 30.641 2.404 .101 Cumulative (Combined) Transpiration Within Groups 229.417 18 12.745 (mL) * Compound Total 321.341 21
173 Carmen G. Franks - Appendices
Table A2.2. Salix ANOVA comparison among compounds for cumulative transpiration volume at each sampling time. Table A2.3 for subsequent analysis.
Time Sum of Mean df F Sig. (hours) Squares Square Cumulative Between Groups 2.000 3 .667 5.333 .008 Transpiration (Combined) 2 (mL) * Within Groups 2.250 18 .125 Compound Total 4.250 21 Between Groups Cumulative 4.975 3 1.658 3.534 .036 Transpiration (Combined) 4 (mL) * Within Groups 8.448 18 .469 Compound Total 13.423 21 Between Groups Cumulative 22.663 3 7.554 3.266 .045 Transpiration (Combined) 8 (mL) * Within Groups 41.635 18 2.313 Compound Total 64.298 21 Between Groups Cumulative 91.924 3 30.641 2.404 .101 Transpiration (Combined) 24 (mL) * Within Groups 229.417 18 12.745 Compound Total 321.341 21
174 Carmen G. Franks - Appendices
Table A2.3. Dunnett C Post Hoc analysis for Salix exigua of cumulative transpiration between compounds and over time ANOVA analysis in Table A2.2 Time (I) (J) Mean Diff. 95% Confidence Interval Std. Error (hrs) Cmpd. Cmpd. (I-J) Lower Bound Upper Bound 2 ATZ DTZ .8333(*) .16667 .2183 1.4483 DZP .1667 .26874 -.8250 1.1583 EE2 .5000 .17480 -.1450 1.1450 DTZ ATZ -.8333(*) .16667 -1.4483 -.2183 DZP -.6667 .21082 -1.4446 .1112 EE2 -.3333(*) .05270 -.5278 -.1389 DZP ATZ -.1667 .26874 -1.1583 .8250 DTZ .6667 .21082 -.1112 1.4446 EE2 .3333 .21731 -.4685 1.1352 EE2 ATZ -.5000 .17480 -1.1450 .1450 DTZ .3333(*) .05270 .1389 .5278 DZP -.3333 .21731 -1.1352 .4685 4 ATZ DTZ 1.2917(*) .32543 .0363 2.5470 DZP .3750 .50861 -1.5017 2.2517 EE2 .9167 .32702 -.2900 2.1234 DTZ ATZ -1.2917(*) .32543 -2.5470 -.0363 DZP -.9167 .42898 -2.5410 .7076 EE2 -.3750 .17970 -1.1368 .3868 DZP ATZ -.3750 .50861 -2.2517 1.5017 DTZ .9167 .42898 -.7076 2.5410 EE2 .5417 .43020 -1.0457 2.1291 EE2 ATZ -.9167 .32702 -2.1234 .2900 DTZ .3750 .17970 -.3868 1.1368 DZP -.5417 .43020 -2.1291 1.0457 8 ATZ DTZ 2.5000(*) .61237 .0472 4.9528 DZP .5417 1.13361 -3.6413 4.7246 EE2 2.0833 .57615 -.0426 4.2093 DTZ ATZ -2.5000(*) .61237 -4.9528 -.0472 DZP -1.9583 1.05755 -5.9725 2.0558 EE2 -.4167 .40654 -2.2078 1.3745 DZP ATZ -.5417 1.13361 -4.7246 3.6413 DTZ 1.9583 1.05755 -2.0558 5.9725 EE2 1.5417 1.03699 -2.2847 5.3681 EE2 ATZ -2.0833 .57615 -4.2093 .0426 DTZ .4167 .40654 -1.3745 2.2078 DZP -1.5417 1.03699 -5.3681 2.2847 24 ATZ DTZ 5.0000 1.31339 -.0265 10.0265 DZP .5833 2.68845 -9.3368 10.5035 EE2 3.8333 1.36117 -1.1893 8.8559 DTZ ATZ -5.0000 1.31339 -10.0265 .0265 DZP -4.4167 2.43299 -13.4914 4.6581 EE2 -1.1667 .73786 -4.2100 1.8767 DZP ATZ -.5833 2.68845 -10.5035 9.3368 DTZ 4.4167 2.43299 -4.6581 13.4914 EE2 3.2500 2.45911 -5.8239 12.3239 EE2 ATZ -3.8333 1.36117 -8.8559 1.1893 DTZ 1.1667 .73786 -1.8767 4.2100 DZP -3.2500 2.45911 -12.3239 5.8239 * The mean difference is significant at the .05 level.
175 Carmen G. Franks - Appendices
Table A2.4. Spearman’s rho correlation analysis for Salix exigua between uptake and either fresh weights (root, shoot, total) or cumulative transpiration volumes over the 24 hour study period.
Shoot Cumulative Time Root fr. Total fr. Compound fr. wt. Transpiration (hours) wt. (g) wt. (g) (g) (mL)
ATZ 2 Corr. Coe. .143 .429 -.029 -.203 Sig. (2-tailed) .787 .397 .957 .700 4 Corr. Coe. -.143 .086 -.314 -.371 Sig. (2-tailed) .787 .872 .544 .468
8 Corr. Coe. .429 .429 -.086 -.086
Sig. (2-tailed) .397 .397 .872 .872 24 Corr. Coe. .886(*) .943(**) .771 .943(**) Sig. (2-tailed) .019 .005 .072 .005
DTZ 2 Corr. Coe. 1.000(**) .600 .600 . Sig. (2-tailed) . .400 .400 . 4 Corr. Coe. .800 .800 .800 .775 Sig. (2-tailed) .200 .200 .200 .225 8 Corr. Coe. .800 .800 .800 1.000(**) Sig. (2-tailed) .200 .200 .200 . 24 Corr. Coe. 1.000(**) .600 .600 .800 Sig. (2-tailed) . .400 .400 .200
DZP 2 Corr. Coe. .371 .086 .116 -.213 Sig. (2-tailed) .468 .872 .827 .686 4 Corr. Coe. .087 -.116 -.088 -.265 Sig. (2-tailed) .870 .827 .868 .612 8 Corr. Coe. .754 .986(**) .956(**) .899(*) Sig. (2-tailed) .084 .000 .003 .015 24 Corr. Coe. .600 .886(*) .812(*) .943(**) Sig. (2-tailed) .208 .019 .050 .005
EE2 2 Corr. Coe. .725 .714 .771 .414 Sig. (2-tailed) .103 .111 .072 .414 4 Corr. Coe. .812(*) .600 .714 .759 Sig. (2-tailed) .050 .208 .111 .080 8 Corr. Coe. .754 .657 .771 .551 Sig. (2-tailed) .084 .156 .072 .257 24 Corr. Coe. .882(*) .638 .754 .500 Sig. (2-tailed) .020 .173 .084 .312 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).
176 Carmen G. Franks - Appendices
Table A2.5. Spearman’s rho correlation for Salix exigua between cumulative transpiration and fresh weights across the study period.
Time Root fr. Shoot fr. Total fr. Compound (hours) wt. (g) wt. (g) wt. (g)
ATZ 2 Corr. Coe. .522 .551 .899(*) Sig. (2-tailed) .288 .257 .015 4 Corr. Coe. .543 .771 .943(**) Sig. (2-tailed) .266 .072 .005 8 Corr. Coe. .714 .829(*) 1.000(**) Sig. (2-tailed) .111 .042 . 24 Corr. Coe. .943(**) .886(*) .886(*) Sig. (2-tailed) .005 .019 .019
DTZa 2 Corr. Coe. . . . Sig. (2-tailed) . . . 4 Corr. Coe. .258 .775 .775 Sig. (2-tailed) .742 .225 .225 8 Corr. Coe. .800 .800 .800 Sig. (2-tailed) .200 .200 .200 24 Corr. Coe. .800 .800 .800 Sig. (2-tailed) .200 .200 .200
DZP 2 Corr. Coe. .638 .941(**) .893(*) Sig. (2-tailed) .173 .005 .016 4 Corr. Coe. .812(*) .986(**) .971(**) Sig. (2-tailed) .050 .000 .001 8 Corr. Coe. .771 .943(**) .928(**) Sig. (2-tailed) .072 .005 .008 24 Corr. Coe. .771 .943(**) .899(*) Sig. (2-tailed) .072 .005 .015
EE2 2 Corr. Coe. .630 .828(*) .828(*) Sig. (2-tailed) .180 .042 .042 4 Corr. Coe. .739 .941(**) .880(*) Sig. (2-tailed) .093 .005 .021 8 Corr. Coe. .647 .986(**) .899(*) Sig. (2-tailed) .165 .000 .015 24 Corr. Coe. .647 .986(**) .899(*) Sig. (2-tailed) .165 .000 .015 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed). a Due to the small sample size for DTZ, the correlation analysis could not be done. See Figure A2.1.
177 Carmen G. Franks - Appendices
Figure A2.1. Scatter plots of DTZ treated plants with plant shoot, root and total fresh weight for each sampling time.
Time (hours): 2 Time (hours): 4
2.00 2.00
R Sq Linear = 0.374
1.50 1.50
1.00 1.00
R Sq Linear = 0.375
0.50 0.50
R Sq Linear = 0.08
0.00 0.00
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.70 1.80 1.90 2.00 2.10 2.20 2.30
Time (hours): 8 Time (hours): 24
2.00 2.00 Fresh weight (g) weight Fresh
R Sq Linear = 0.797 R Sq Linear = 0.897 1.50 1.50
1.00 1.00
R Sq Linear = 0.745 R Sq Linear = 0.853
0.50 0.50
R Sq Linear = 0.727 R Sq Linear = 0.62
0.00 0.00
3.25 3.50 3.75 4.00 4.25 4.50 4.75 8.00 8.50 9.00 9.50 10.00 10.50 Cumulative transpiration volume (mL) Root fr. wt. (g) Cumulative Transpiration (mL) Shoot fr. wt. (g) Cumulative Transpiration (mL) Total fr. wt. (g) Cumulative Transpiration (mL)
178 Carmen G. Franks - Appendices
Table A2.6. ANOVA analysis root fresh weight for Salix exigua RCF plants among type, excised roots and roots attached to the cutting.
Sum of Mean df F Sig. Squares Square ATZ Root fr. wt. Between Groups .037 1 .037 .925 .391 (g) * Type (Combined) Within Groups .159 4 .040 Total .196 5 DZP Root fr. wt. Between Groups .035 2 .017 .224 .812 (g) * Type (Combined) Within Groups .232 3 .077 Total .267 5 EE2 Root fr. wt. Between Groups .002 1 .002 .158 .711 (g) * Type (Combined) Within Groups .061 4 .015 Total .063 5
179 Carmen G. Franks - Appendices
Table A2.7. ANOVA analysis of root uptake in experiments with shoots removed from Salix exigua between excised roots and roots+wood for ATZ, DZP and EE2. Type, roots excised or roots attached to wood stem. Time Sum of Mean (hrs) Squares df Square F Sig. ATZ 1 Percent Uptake * Between (Combined) 193.802 1 193.802 1.356 .309 Type Groups Within Groups 571.633 4 142.908 Total 765.435 5 2 Percent Uptake * Between (Combined) 1284.807 1 1284.807 9.214 .039 Type Groups Within Groups 557.753 4 139.438 Total 1842.560 5 4 Percent Uptake * Between (Combined) 184.815 1 184.815 1.349 .310 Type Groups Within Groups 547.953 4 136.988 Total 732.768 5 8 Percent Uptake * Between (Combined) 218.407 1 218.407 1.776 .253 Type Groups Within Groups 491.907 4 122.977 Total 710.313 5 24 Percent Uptake * Between (Combined) 147.015 1 147.015 1.239 .328 Type Groups Within Groups 474.800 4 118.700 Total 621.815 5 DZP 1 Percent Uptake * Between (Combined) 421.682 1 421.682 11.948 .026 Type Groups Within Groups 141.167 4 35.292 Total 562.848 5 2 Percent Uptake * Between (Combined) 285.660 1 285.660 13.712 .021 Type Groups Within Groups 83.333 4 20.833 Total 368.993 5 4 Percent Uptake * Between (Combined) 245.760 1 245.760 6.415 .064 Type Groups Within Groups 153.233 4 38.308 Total 398.993 5 8 Percent Uptake * Between (Combined) 308.167 1 308.167 36.078 .004 Type Groups Within Groups 34.167 4 8.542 Total 342.333 5 24 Percent Uptake * Between (Combined) 220.827 1 220.827 6.695 .061 Type Groups Within Groups 131.933 4 32.983 Total 352.760 5 EE2 1 Percent Uptake * Between (Combined) 1561.707 1 1561.707 10.935 .030 Type Groups Within Groups 571.253 4 142.813 Total 2132.960 5 2 Percent Uptake * Between (Combined) 38.002 1 38.002 .953 .384 Type Groups Within Groups 159.507 4 39.877 Total 197.508 5 4 Percent Uptake * Between (Combined) 1053.375 1 1053.375 7.950 .048 Type Groups Within Groups 529.973 4 132.493 Total 1583.348 5 8 Percent Uptake * Between (Combined) 622.202 1 622.202 10.657 .031 Type Groups Within Groups 233.547 4 58.387 Total 855.748 5 24 Percent Uptake * Between (Combined) 143.082 1 143.082 3.068 .155 Type Groups Within Groups 186.567 4 46.642 Total 329.648 5
180 Carmen G. Franks - Appendices
Table A2.8. ANOVA analysis of root uptake in experiments with shoots removed from Salix exigua among compounds over time. Subsequent Tamhane post hoc analysis in Figure A2.9.
Time Sum of Mean df F Sig. (hrs) Squares Square
Between Groups 417.528 2 208.764 .905 .426 Percent (Combined) 1 Uptake * Within Groups 3461.243 15 230.750 Compound Total 3878.771 17
Between Groups 2611.963 2 1305.982 8.132 .004 Percent (Combined) 2 Uptake * Within Groups 2409.062 15 160.604 Compound Total 5021.025 17
Between Groups 2813.521 2 1406.761 7.772 .005 Percent (Combined) 4 Uptake * Within Groups 2715.110 15 181.007 Compound Total 5528.631 17
Between Groups 5032.588 2 2516.294 19.778 .000 Percent (Combined) 8 Uptake * Within Groups 1908.395 15 127.226 Compound Total 6940.983 17
Between Groups 7408.528 2 3704.264 42.603 .000 Percent (Combined) 24 Uptake * Within Groups 1304.223 15 86.948 Compound Total 8712.751 17
181 Carmen G. Franks - Appendices
Table A2.9. Tamhane post hoc analysis for ANOVA analysis of root uptake in experiments with shoots removed from Salix exigua among compounds presented in Table A2.7.
Time (I) (J) Mean Diff. Std. 95% Confidence Interval Sig. (hrs) Cmpd. Cmpd. (I-J) Error Lower Bound Upper Bound 1 ATZ DZP -1.167 6.6540 .998 -20.284 17.951 EE2 -10.750 9.8292 .665 -40.121 18.621 DZP ATZ 1.167 6.6540 .998 -17.951 20.284 EE2 -9.583 9.4795 .717 -38.582 19.416 EE2 ATZ 10.750 9.8292 .665 -18.621 40.121 DZP 9.583 9.4795 .717 -19.416 38.582 2 ATZ DZP 11.033 8.5859 .561 -15.800 37.867 EE2 -18.183 8.2463 .194 -45.087 8.720 DZP ATZ -11.033 8.5859 .561 -37.867 15.800 EE2 -29.217(*) 4.3455 .000 -41.870 -16.564 EE2 ATZ 18.183 8.2463 .194 -8.720 45.087 DZP 29.217(*) 4.3455 .000 16.564 41.870 4 ATZ DZP -5.383 6.1421 .787 -23.251 12.484 EE2 -28.800(*) 8.7866 .029 -54.603 -2.997 DZP ATZ 5.383 6.1421 .787 -12.484 23.251 EE2 -23.417 8.1288 .066 -48.374 1.540 EE2 ATZ 28.800(*) 8.7866 .029 2.997 54.603 DZP 23.417 8.1288 .066 -1.540 48.374 8 ATZ DZP -5.200 5.9235 .787 -22.552 12.152 EE2 -37.783(*) 7.2251 .001 -58.483 -17.084 DZP ATZ 5.200 5.9235 .787 -12.152 22.552 EE2 -32.583(*) 6.3195 .002 -51.321 -13.846 EE2 ATZ 37.783(*) 7.2251 .001 17.084 58.483 DZP 32.583(*) 6.3195 .002 13.846 51.321 24 ATZ DZP -12.250 5.6996 .167 -28.796 4.296 EE2 -47.833(*) 5.6316 .000 -64.240 -31.427 DZP ATZ 12.250 5.6996 .167 -4.296 28.796 EE2 -35.583(*) 4.7694 .000 -49.228 -21.939 EE2 ATZ 47.833(*) 5.6316 .000 31.427 64.240 DZP 35.583(*) 4.7694 .000 21.939 49.228 * The mean difference is significant at the .05 level.
182 Carmen G. Franks - Appendices
Chapter 3
Table A3.1. Arabidopsis ANOVA analysis among compounds for averaged cumulative transpiration volume, root, shoot and total fresh weight of replicate plants.
Sum of Mean df F Sig. Squares Square Between Groups .059 3 .020 .250 .860 (Combined) Root fr. wt. (g) * Compound Within Groups 1.802 23 .078 Total 1.861 26 Between Groups .567 3 .189 .638 .598 (Combined) Shoot fr. wt. (g) * Compound Within Groups 6.814 23 .296 Total 7.382 26 Between Groups .873 3 .291 .473 .704 Total fr. wt. (g) * (Combined) Compound Within Groups 14.160 23 .616 Total 15.033 26 Between Groups 12.700 3 4.233 .980 .420 Cumulative (Combined) Transpiration (mL) * Within Groups 99.402 23 4.322 Compound Total 112.102 26
183 Carmen G. Franks - Appendices
Table A3.2. Arabidopsis ANOVA comparison among the different compounds of the mean transpired volume at each sampling time. Subsequent LSD post hoc analysis presented in Table A3.3.
Time Sum of Mean df F Sig. (hours) Squares Square Cumulative Between Groups 1.231 3 .410 6.576 .002 Transpiration (Combined) 2 (mL) * Within Groups 1.435 23 .062 Compound Total 2.667 26 Between Groups Cumulative 2.941 3 .980 6.533 .002 Transpiration (Combined) 4 (mL) * Within Groups 3.452 23 .150 Compound Total 6.394 26 Between Groups Cumulative 3.832 3 1.277 .810 .501 Transpiration (Combined) 10 (mL) * Within Groups 36.256 23 1.576 Compound Total 40.088 26 Between Groups Cumulative 12.700 3 4.233 .980 .420 Transpiration (Combined) 24 (mL) * Within Groups 99.402 23 4.322 Compound Total 112.102 26
184 Carmen G. Franks - Appendices
Table A3.3. LSD post hoc analysis for Arabidopsis cumulative transpiration among compounds for each sampling time. Extension of Table A3.2. Time (I) (J) Mean Diff. Std. 95% Confidence Interval Sig. (hrs) Cmpd. Cmpd. (I-J) Error Lower Bound Upper Bound 2 ATZ DTZ .3167(*) .15127 .048 .0037 .6296 DZP .3583(*) .15127 .027 .0454 .6713 EE2 .6000(*) .13683 .000 .3169 .8831 DTZ ATZ -.3167(*) .15127 .048 -.6296 -.0037 DZP .0417 .14423 .775 -.2567 .3400 EE2 .2833(*) .12901 .038 .0165 .5502 DZP ATZ -.3583(*) .15127 .027 -.6713 -.0454 DTZ -.0417 .14423 .775 -.3400 .2567 EE2 .2417 .12901 .074 -.0252 .5085 EE2 ATZ -.6000(*) .13683 .000 -.8831 -.3169 DTZ -.2833(*) .12901 .038 -.5502 -.0165 DZP -.2417 .12901 .074 -.5085 .0252 4 ATZ DTZ .4250 .23459 .083 -.0603 .9103 DZP .2167 .23459 .365 -.2686 .7020 EE2 .8500(*) .21220 .001 .4110 1.2890 DTZ ATZ -.4250 .23459 .083 -.9103 .0603 DZP -.2083 .22367 .361 -.6710 .2544 EE2 .4250(*) .20006 .045 .0111 .8389 DZP ATZ -.2167 .23459 .365 -.7020 .2686 DTZ .2083 .22367 .361 -.2544 .6710 EE2 .6333(*) .20006 .004 .2195 1.0472 EE2 ATZ -.8500(*) .21220 .001 -1.2890 -.4110 DTZ -.4250(*) .20006 .045 -.8389 -.0111 DZP -.6333(*) .20006 .004 -1.0472 -.2195 10 ATZ DTZ 1.0000 .76026 .201 -.5727 2.5727 DZP 1.0000 .76026 .201 -.5727 2.5727 EE2 .9250 .68768 .192 -.4976 2.3476 DTZ ATZ -1.0000 .76026 .201 -2.5727 .5727 DZP .0000 .72488 1.000 -1.4995 1.4995 EE2 -.0750 .64835 .909 -1.4162 1.2662 DZP ATZ -1.0000 .76026 .201 -2.5727 .5727 DTZ .0000 .72488 1.000 -1.4995 1.4995 EE2 -.0750 .64835 .909 -1.4162 1.2662 EE2 ATZ -.9250 .68768 .192 -2.3476 .4976 DTZ .0750 .64835 .909 -1.2662 1.4162 DZP .0750 .64835 .909 -1.2662 1.4162 24 ATZ DTZ 1.7250 1.25884 .184 -.8791 4.3291 DZP 2.0167 1.25884 .123 -.5874 4.6208 EE2 1.2000 1.13866 .303 -1.1555 3.5555 DTZ ATZ -1.7250 1.25884 .184 -4.3291 .8791 DZP .2917 1.20025 .810 -2.1912 2.7746 EE2 -.5250 1.07354 .629 -2.7458 1.6958 DZP ATZ -2.0167 1.25884 .123 -4.6208 .5874 DTZ -.2917 1.20025 .810 -2.7746 2.1912 EE2 -.8167 1.07354 .455 -3.0375 1.4041 EE2 ATZ -1.2000 1.13866 .303 -3.5555 1.1555 DTZ .5250 1.07354 .629 -1.6958 2.7458 DZP .8167 1.07354 .455 -1.4041 3.0375
185 Carmen G. Franks - Appendices
Table A3.4. Correlation analysis between uptake from solution, plant fresh weight (root, shoot and total) and cumulative transpired volume, at each sampling time. Spearman’s rho.
Time Root fr. Shoot fr. Total fr. Cum.
(hours) wt. (g) wt. (g) wt. (g) Trans.(mL)
ATZ 2 Correlation Co. .667 -.300 .700 .527 Sig. (2-tailed) .219 .624 .188 .361 4 Correlation Co. .667 -.300 .700 .527 Sig. (2-tailed) .219 .624 .188 .361
10 Correlation Co. .667 .100 .900* .800 Sig. (2-tailed) .219 .873 .037 .104 24 Correlation Co. .667 .100 .900* .800 Sig. (2-tailed) .219 .873 .037 .104
DTZ 2 Correlation Co. .086 -.143 -.143 -.131 Sig. (2-tailed) .872 .787 .787 .805 4 Correlation Co. -.371 -.429 -.429 .507 Sig. (2-tailed) .468 .397 .397 .305
10 Correlation Co. .029 -.086 -.086 -.319 Sig. (2-tailed) .957 .872 .872 .538 24 Correlation Co. -.143 -.200 -.200 -.657 Sig. (2-tailed) .787 .704 .704 .156
DZP 2 Correlation Co. .257 .143 .086 .655 Sig. (2-tailed) .623 .787 .872 .158 4 Correlation Co. .371 .486 .314 -.638 Sig. (2-tailed) .468 .329 .544 .173
10 Correlation Co. -.257 .086 -.086 -.429 Sig. (2-tailed) .623 .872 .872 .397 24 Correlation Co. -.200 -.086 -.143 -.143 Sig. (2-tailed) .704 .872 .787 .787
EE2 2 Correlation Co. .721* .673* .685* .436 Sig. (2-tailed) .019 .033 .029 .208 4 Correlation Co. .491 .418 .442 .430 Sig. (2-tailed) .150 .229 .200 .215
10 Correlation Co. .552 .467 .491 .245 Sig. (2-tailed) .098 .174 .150 .496 24 Correlation Co. .588 .539 .552 .535 Sig. (2-tailed) .074 .108 .098 .111
* Correlation is significant at the 0.05 level (2-tailed).
186 Carmen G. Franks - Appendices
Table A3.5. Arabidopsis correlation analysis of cumulative transpired volume at each sampling time correlated to plant fresh weights (root, shoot and total). Spearman’s rho. Time Root fr. Shoot fr. Total fr.
(hours) wt. (g) wt. (g) wt. (g)
ATZ 2 Correlation Co. .108 -.369 .316 Sig. (2-tailed) .863 .541 .604 4 Correlation Co. .108 -.369 .316 Sig. (2-tailed) .863 .541 .604
10 Correlation Co. .564 .400 .900* Sig. (2-tailed) .322 .505 .037 24 Correlation Co. .564 .400 .900* Sig. (2-tailed) .322 .505 .037
DTZ 2 Correlation Co. -.393 -.393 -.393 Sig. (2-tailed) .441 .441 .441 4 Correlation Co. .541 .541 .541 Sig. (2-tailed) .268 .268 .268
10 Correlation Co. .783 .725 .725 Sig. (2-tailed) .066 .103 .103 24 Correlation Co. .771 .829* .829* Sig. (2-tailed) .072 .042 .042
DZP 2 Correlation Co. .655 .393 .393 Sig. (2-tailed) .158 .441 .441 4 Correlation Co. .232 .290 .464 Sig. (2-tailed) .658 .577 .354
10 Correlation Co. .314 .371 .543 Sig. (2-tailed) .544 .468 .266 24 Correlation Co. .486 .600 .714 Sig. (2-tailed) .329 .208 .111
EE2 2 Correlation Co. .658* .658* .703* Sig. (2-tailed) .039 .039 .023 4 Correlation Co. .823** .898** .860** Sig. (2-tailed) .003 .000 .001
10 Correlation Co. .826** .924** .887** Sig. (2-tailed) .003 .000 .001 24 Correlation Co. .906** .967** .942** Sig. (2-tailed) .000 .000 .000
* Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).
187 Carmen G. Franks - Appendices
Chapter 5
Table A5.1. ANOVA comparison of root and shoot fresh weight for Arabidopsis and willow among compounds.
Sum of Mean Cmpd df F Sig. Squares Square
ATZ Root * Between Groups .101 1 .101 2.106 .181 Plant (Combined) Within Groups .431 9 .048 Total .532 10 Shoot * Between Groups 1.549 1 1.549 29.857 .000 Plant (Combined) Within Groups .467 9 .052 Total 2.016 10
DTZ Root * Between Groups .241 1 .241 5.458 .048 Plant (Combined) Within Groups .353 8 .044 Total .593 9 Shoot * Between Groups 2.477 1 2.477 26.919 .001 Plant (Combined) Within Groups .736 8 .092 Total 3.213 9
DZP Root * Between Groups .078 1 .078 1.594 .235 Plant (Combined) Within Groups .492 10 .049 Total .570 11 Shoot * Between Groups .859 1 .859 5.340 .043 Plant (Combined) Within Groups 1.608 10 .161 Total 2.467 11
EE2 Root * Between Groups .380 1 .380 4.179 .060 Plant (Combined) Within Groups 1.273 14 .091 Total 1.653 15 Shoot * Between Groups 4.871 1 4.871 13.261 .003 Plant (Combined) Within Groups 5.142 14 .367 Total 10.013 15
188 Carmen G. Franks - Appendices
Table A5.2. ANOVA comparison of total volume transpired among Arabidopsis and willow for the four compounds from the 24 hours uptake studies.
Sum of Mean df F Sig. Squares Square
ATZ CumTrans Between Groups 31.527 1 31.527 4.636 .060 * Plant (Combined) Within Groups 61.200 9 6.800 Total 92.727 10 DTZ CumTrans Between Groups .038 1 .038 .018 .897 * Plant (Combined) Within Groups 16.719 8 2.090 Total 16.756 9 DZP CumTrans Between Groups 70.083 1 70.083 3.739 .082 * Plant (Combined) Within Groups 187.417 10 18.742 Total 257.500 11 EE2 CumTrans Between Groups 2.204 1 2.204 .486 .497 * Plant (Combined) Within Groups 63.483 14 4.535 Total 65.687 15
189 Carmen G. Franks - Appendices
Table A5.3a. ANOVA comparison for percent uptake between Arabidopsis and willow across each sampling time for 17α-ethynylestradiol. Note that sampling time for willow was 0, 2, 4, 8, 24 and Arabidopsis was 0, 2, 4, 10, 24, therefore times t = 8 and t = 10 are considered the same sampling time for these comparisons.
Sum of Mean Time df F Sig. Squares Square
17α- 2.00 Uptake * Between Groups 2478.958 1 2478.958 39.795 .000 ethynylestradiol Plant (Combined) Within Groups 872.114 14 62.294 Total 3351.072 15 4.00 Uptake * Between Groups 641.280 1 641.280 10.738 .006 Plant (Combined) Within Groups 836.049 14 59.718 Total 1477.329 15 8.00 Uptake * Between Groups 139.416 1 139.416 2.680 .124 Plant (Combined) Within Groups 728.338 14 52.024 Total 867.754 15 24.00 Uptake * Between Groups 40.788 1 40.788 2.689 .123 Plant (Combined) Within Groups 212.332 14 15.167 Total 253.120 15
190 Carmen G. Franks - Appendices
Table A5.3b. ANOVA comparison for percent uptake between Arabidopsis and willow across each sampling time for diltiazem. Note that sampling time for willow was 0, 2, 4, 8, 24 and Arabidopsis was 0, 2, 4, 10, 24, therefore times t = 8 and t = 10 are considered the same sampling time for these comparisons.
Sum of Mean Time df F Sig. Squares Square diltiazem 2.00 Uptake * Between Groups 1107.165 1 1107.165 7.468 .026 Plant (Combined) Within Groups 1186.008 8 148.251 Total 2293.174 9 4.00 Uptake * Between Groups 1554.282 1 1554.282 15.521 .004 Plant (Combined) Within Groups 801.135 8 100.142 Total 2355.417 9 8.00 Uptake * Between Groups 2551.928 1 2551.928 26.128 .001 Plant (Combined) Within Groups 781.349 8 97.669 Total 3333.277 9 24.00 Uptake * Between Groups 996.664 1 996.664 15.777 .004 Plant (Combined) Within Groups 505.366 8 63.171 Total 1502.030 9
191 Carmen G. Franks - Appendices
Table A5.3c. ANOVA comparison for percent uptake between Arabidopsis and willow across each sampling time for diazepam. Note that sampling time for willow was 0, 2, 4, 8, 24 and Arabidopsis was 0, 2, 4, 10, 24, therefore times t = 8 and t = 10 are considered the same sampling time for these comparisons.
Sum of Mean Time df F Sig. Squares Square diazepam 2.00 Uptake * Between Groups 143.037 1 143.037 3.422 .094 Plant (Combined) Within Groups 417.994 10 41.799 Total 561.031 11 4.00 Uptake * Between Groups 34.544 1 34.544 1.002 .340 Plant (Combined) Within Groups 344.753 10 34.475 Total 379.297 11 8.00 Uptake * Between Groups 300.400 1 300.400 5.882 .036 Plant (Combined) Within Groups 510.712 10 51.071 Total 811.112 11 24.00 Uptake * Between Groups 17.017 1 17.017 .289 .603 Plant (Combined) Within Groups 589.079 10 58.908 Total 606.096 11
192 Carmen G. Franks - Appendices
Table A5.3d. ANOVA comparison for percent uptake among Arabidopsis and willow across each sampling time for atrazine. Note that sampling time for willow was 0, 2, 4, 8, 24 and Arabidopsis was 0, 2, 4, 10, 24, therefore times t = 8 and t = 10 are considered the same sampling time for these comparisons.
Sum of Mean Time df F Sig. Squares Square atrazine 2.00 Uptake * Between Groups 398.992 1 398.992 11.070 .009 Plant (Combined) Within Groups 324.390 9 36.043 Total 723.382 10 4.00 Uptake * Between Groups 47.530 1 47.530 2.559 .144 Plant (Combined) Within Groups 167.169 9 18.574 Total 214.699 10 8.00 Uptake * Between Groups 29.305 1 29.305 .311 .591 Plant (Combined) Within Groups 848.619 9 94.291 Total 877.925 10 24.00 Uptake * Between Groups 9.598 1 9.598 .133 .724 Plant (Combined) Within Groups 648.682 9 72.076 Total 658.280 10
193 Carmen G. Franks - Appendices
APPENDIX B Other experimental methods
194 Carmen G. Franks - Appendices
Preliminary investigation: Salix exigua
Introduction. An initial investigation involving EE2 and several willow plants examined the effect of different concentrations, as well as uptake time, on the distribution of the EE2 with willow. As the number of plants is small, n = 1, the results presented here have not been affirmed, but may still suggest the effect of concentration and time on uptake and distribution of neutral pharmaceuticals.
Materials and methods. Three concentrations of unlabeled EE2 were added to three culture tubes, 0, 1, and 10 ug, with 16,667 Bq of 3H-EE2 and 24 mL of hydroponic nutrient solution. Willows were inserted into each culture tube (with the cutting maintained out of the solution), the tubes covered with foil and harvested after 8 hours. The roots were rinsed with nutrient solution, the plants separated into roots, shoot and cutting, weighed fresh, and stored at -20°C until time of solvent extraction and oxidation.
As well, two other culture tubes were set up with 0 ug unlabeled EE2, 16,667 Bq 3H-EE2, and 24 mL nutrient solution. Willows were inserted into the tubes and left for 24 and 48 hours before harvesting. At experiment termination, the roots were rinsed with nutrient solution and the plants separated into components, weighed fresh, and stored at -20°C until time of analysis.
Plant components were analyzed as described in section 2.2.5 and 2.2.6, including soluble extraction, and oxidation of residues for bound fractions and only oxidation of the cutting.
Results and discussion. Results from the preliminary investigation with EE2, although not replicated, still present useful information concerning what happens over time and uptake of different concentrations. Levels of recovery vary between the plants, but a general trend can still be extrapolated.
Uptake trends and values for the different concentrations were similar, except for the 8 hour 0 μg EE2 value (Figure B1.1).
Distribution of recovered radioactivity for the 3 plants exposed to 3 different concentrations of EE2 and harvested after 8 hours appear to be similar (Figure B1.2). It appears though, that lower concentrations are taken up at a faster rate than higher concentrations. Proportions of root soluble 3H-EE2 appear to increase with decreasing concentrations. If uptake and distribution varies with concentration, it does not appear to be significant with these results.
Distribution of recovered radioactivity for the 3 plants exposed to one concentration of EE2 (0 ug + 16,667 Bq 3H-EE2) and one plant harvested at 8, 24, and 48 hours clearly show differences (Figure B1.3). At 8 hours there is a considerably higher proportion of root soluble 3H-EE2 than at 24 and 48 hours, with 48% versus 7% for both 24 and 48 hours. By 24 hours, the proportion of root bound 3H-EE2 has increased to 75% and at 48
195 Carmen G. Franks - Appendices hours 98%, compared to the 48% at 8 hours. Proportions within the shoot are close to the same, and therefore difficult to determine if there are differences over time.
Conclusions. Considering the data presented, there appears to be little effect of concentration on uptake and distribution for EE2 and willow, but a definite effect of time on distribution within the plants. With increased in-plant time, the proportion of compound becoming bound to the root also increased (with the coinciding decrease in soluble compound). Although the results have not been replicated, they do provide suggestions into the effect of concentration and time on the uptake and distribution of EE2, and possibly other neutral pharmaceuticals, within Salix exigua.
196 Carmen G. Franks - Appendices
100
75
50 Uptake (%)
25
0 ug EE2 1ug EE2 10ug EE2
0 0 8 16 24 32 40 48 Time (hours)
Figure B1.1. Uptake of 3 concentrations of 17α-ethynylestradiol (EE2) (0, 1, and 10 μg EE2 + 16,667 Bq 3H-EE2) from 24 mL of solution. 0 μg EE2, n = 3, one plant harvested at 8, 24 and 48 hours; 1 μg EE2, n = 1, plant harvested at 8 hours; 10 μg EE2, n = 1, solution sampled at 2, 4, 6, and 8 hours with harvest at 8 hours.
197 Carmen G. Franks - Appendices
0 ug 1 ug 10 ug 100% )
50% Distribution of recovered radioactivity (%
0% Plant 1 Plant 2 Plant 3
Wood, Whole, Oxidized 14.2 6.7 5.8 Shoot Bound, Oxidized Residue 6.5 1.6 1.7 Shoot Soluble, Solvent Extract 6.6 0.8 1.8 Root Bound, Oxidized Residue 48.3 37.8 36.7 Root Soluble, Solvent Extract 48.1 23.9 19.8
Figure B1.2. Distribution of recovered radioactivity from 3 Salix exigua cuttings exposed to 16,667 Bq 3H-17α-ethynylestradiol and 3 concentrations of 17α- ethynylestradiol (EE2); 0, 1 and 10 μg, in 24 mL of nutrient solution. All three plants were harvested after 8 hours. Recovery values are presented in the table.
198 Carmen G. Franks - Appendices
8 hours 24 hours 48 hours
) 100%
50% Distribution of recovered radioactivity (% 0% Plant 1 Plant 2 Plant 3
Wood, Whole, Oxidized 14.2 n/a0 4.0 Shoot Bound, Oxidized Residue 6.5 2.7 2.0 Shoot Soluble, Solvent Extract 6.6 3.3 3.1 Root Bound, Oxidized Residue 48.3 75.0 97.7 Root Soluble, Solvent Extract 48.1 6.6 7.2
Figure B1.3. Distribution of recovered radioactivity from 3 Salix exigua cuttings exposed to one concentration of 17α-ethynylestradiol (EE2) (0 μg) containing 16,667 Bq 3H-17α-ethynylestradiol in 24 mL solution. One plant was harvested at each of 8, 24 and 48 hours. Recovery values are presented in the table. Plant 2 Wood, Whole, Oxidized value is not zero, but oxidation was not performed on the cutting as the cutting had previously undergone solvent extraction and oxidation was not possible.
199 Carmen G. Franks - Appendices
Cell wall fractionation and cell wall dissolution
Several methods were attempted in an effort to release the bound compounds for quantification, but neither achieved the desired results. A process of step-wise cell wall fractionation was adapted from Laurent and Scalla (2000)1 and cell wall dissolution adapted from Lu and Ralph (2003)2 were attempted. Both processes result in samples that are permanently pigmented primarily due to lignin, which interferes with LSC.
Cell wall fractionation. A step-wise procedure to break down the cell wall into 5 constituent parts was attempted. The steps are outlined below in order of procedure, each commencing with a period of sonication prior to heating or cooling. Between each step, neutralizing the solution, filtration through glass fiber discs (GF/D) and rinsing with water were performed. Both the remaining residue and the glass fiber were carried into the next fractionation step to ensure no losses of cell wall particulates. 1. Pectins – water, 4 °C, 1 h + 100 °C, 2 h, then EDTA, 0.5%, 100 °C, 1 h 2. Hemicellulose I – KOH, 4%, 3 x 6 h 3. Lignin – NAClO4 (sodium perchlorate), 1%, 80 °C, 3 x 3 h 4. Hemicellulose II – KOH, 24%, 3 x 6 h 5. Cellulose – H2SO4, 72%, 20 °C, 1 h, then 3%, 120 °C, 1h
This method was not suitable for extracting and measuring bound radioactivity. Removal of fractions involving an acid or base resulted in pigmented samples. The transfer of residue and glass fiber into the next step resulted in a large mass of glass fiber by the end of the procedure which made it difficult to ensure a sufficient volume was being used to rinse the fraction of interest from the fiber (also resulted in large volumes of solution with small amounts of radioactivity). As well, there was no way to ensure that the glass fiber was successfully preventing the flow of non-step specific cell wall constituents into the collected fraction. Neutralization of acidic and basic fractions with addition of dilute bases and acids also resulted in a large final volume to radioactivity ratio that made it difficult to measure the radioactivity.
Cell wall dissolution. A method combining dimethylsulfoxide (DMSO) and tetrabutylammonium fluoride (TBAF) was reported for use of dissolution of cell walls in the examination of constituents for high-resolution solution-state NMR spectroscopy, resulting in a transparent solution. This method was intriguing as it offered a means of complete dissolution of large particulates that could block tritium beta energy. A ratio of dried, ground residue to volume of DMSO and weight of TBAF was determined to be 225 mg: 6 mL DMSO: 1 g TBAF. The mixture was heated to 60 °C for 30 to 60 min. The result was a transparent, brown solution, with fine particles that readily settled out. This method was not suitable for removal and measuring of bound radioactivity. The coloration was not removable (and interfered with LSC) and the ratio of dried sample to DMSO:TBAF volume was large and therefore not cost effective and could result in a
1 Laurent, F and Scalla, R. 2000. Phenoxyacetic acid residue incorporation in cell walls of soybean (Glycine max.). J. Agric. Food Chem. 48: 4389-4398. 2 Lu, F and Ralph, J. 2003 Non-degredative dissolution and acetylation of ball-milled plant cell walls: high-resolution solution-state NMR. The Plant Journal 35: 535-544.
200 Carmen G. Franks - Appendices small amount of radioactivity within a large volume of solution, making it difficult to measure.
201 Carmen G. Franks - Appendices
SPE of Pigmented Plant Extract Soluble Fractions
Examination of distribution of pharmaceuticals within Arabidopsis initially involved a purification step of pigmented extracts of shoots prior to LSC of the soluble fraction. The pigmented extracts were run through a 1 g – C18 column and stripped with 80 % aqueous methanol (methanol : water 80 : 20). Counts of the purified extracts appeared low. Concern that losses were occurring during this step, an alternative method involving sodium hypochlorite (bleach) was used to create samples that could be reliably analyzed by LSC (method discussed below). Commercially available bleach (5.25 % chlorine) was added to pigmented samples, heated until clear, and then analyzed by LSC. Comparison of recovery results for Arabidopsis DTZ and DZP replicates show marked increase in recovery with the addition of bleach over purification (Figure B1.4).
DTZ Soluble Recovery DZP Soluble Recovery
60
40
20 Recovery (%)
0 C18 no C18 no C18 no C18 no C18
Figure B1.4. Total recovered soluble radioactivity for two methods of sample preparation prior to LSC for DTZ and DZP shoots. C18, soluble fractions underwent C18 purification to remove pigments; no C18, the soluble fractions were bleached to remove pigments. 5 individual examples are shown for DZP and 3 for DTZ.
202 Carmen G. Franks - Appendices
LSC Pigmented Plant Samples – the effects of bleach
To determine the quantity of bleach necessary to achieve full recovery of radioactivity, fresh Arabidopsis leaves were extracted, aliquots taken and radioactive standard added. Bleach was added in increasing amounts and the samples analyzed by LSC. The process and final suggested procedure for bleaching is outlined below.
Methods. Fresh Arabidopsis leaves were solvent extracted as described in section 2.2.5. Following sampling techniques, 2 x 1.5 mL and 2 x 2.0 mL green samples were taken, radiolabeled standard was added to the pigmented extract (2000 Bq of 14C) and counted by LSC.
Another set of aliquots were taken (1.5 mL and 2.0 mL volumes) and radiolabeled standard was added (2000 Bq 14C). Sodium hypochlorite (chlorine bleach, pH 12.42) was used in 5 μL increments from 10 to 35 μL and at 50 μL for the 1.5 mL aliquots and 10, 15, 25 and 30 μL for the 2.0 mL aliquots. The vials capped, shaken, heated to 60 °C for 2 hours and dark adapted prior to LSC analysis.
Results. Recovery of the radioactive standard improved with increasing bleach quantities up to approximately 30 μL (Figure B1.5) for both the 1.5 and 2.0 mL aliquot sizes.
Suggested procedure based on above results:
1. Place 1.5 mL* green sample in scintillation vial. 2. Add 50 μL of sodium hypochlorite solution (5.25% household bleach). 3. Cap tightly, shake, and heat to 50 - 60°C for 1-2 hours (time is based on removal of pigmentation). 4. Cool to room temp and vent chlorine gas into fume hood. Blow out any remaining in the vial with N2 stream. 5. Add 5 mL scintillation cocktail. Shake well. Wipe vial with anti-static sheet. Dark adapt for at least an hour before counting.
The quantity of bleach may vary depending on the degree of pigmentation. It may be necessary to add as much as 200 uL of bleach to create a clear sample, but care must be taken to ensure that phase separation does not occur upon the addition of scintillation cocktail. Bleach quantity may also decrease the pH of the sample past the capabilities of the cocktail. As well, the volume of bleach added may result in chemoluminescence and interfere with counting efficiency that can be noted by changes in recovery of the same sample over time.3
* This volume was found to be a near maximum that could be counted in 7 mL scintillation vials upon the addition of 5 mL cocktail for a sample containing aqueous MeOH with around 6% H2O.
3 Smith, IK and Lang, AL. 1987. Decoloration and solubilization of plant tissue prior to determination of 3H, 14C, and 35S by liquid scintillation. Anal. Biochem. 164: 531-536.
203 Carmen G. Franks - Appendices
Actual dpm (1.5 mL samples) Actual dpm (2 mL samples)
12000
8000
4000 Radioactivity (dpm)
0 0 1020304050 Bleach volume (uL)
Figure B1.5. Effects of bleach on LSC analysis of pigmented plant extracts. Two pigmented sample volumes (1.5 mL and 2 mL) with the addition of approximately 2000 Bq 14C were treated to increasing volumes of bleach (sodium hypochlorite). Bleached samples were heated to 60 °C for 2 hours and allowed to dark adapt prior to LSC analysis.
204 Carmen G. Franks - Appendices
APPENDIX C NMR analysis of 17α-ethynylestradiol
205 Carmen G. Franks - Appendices
Figure C1.1. NMR purity analysis (clean standard) of 17α-ethynylestradiol performed by Dr. P. Dibble.
206 Carmen G. Franks - Appendices
APPENDIX D The pharmaceuticals and herbicide: digging deeper
207 Carmen G. Franks - Appendices
17α-ethynylestradiol. A synthetic estrogen, 17α-ethynylestradiol (EE2), is a common oral contraceptive used in birth control pills. Although the estrogenic effects of EE2 were known in the 1940’s, and it was used widespread as a contraceptive by the 1960’s, the first report on its metabolism in man was not until 1969 (Reed et al., 1972).
Synthetic estrogens are more stable in biological environments, lending to their widespread use. They are not metabolized easily and remain in active form for longer periods of time than natural estrogens.Due to the concerns generated by EE2 and its presence in the environment, research into wastewater and sewage treatment plant methods for degrading this estrogen into less active compounds has increased. Chlorination of EE2 resulted in rapid consumption, but the chlorinated products remained estrogenic and stable for long periods of time in the chlorine environment (Moriyama et al., 2004).
Evidence suggests that some plants have their own endogenous levels of mammalian sex hormones during peak times of reproduction and development. The role of these hormones is not yet understood and neither are the biochemical pathways. If these hormones are produced endogenously and play an important role in flower and embryo development, then is it possible that exogenous hormones, proven to be readily taken up, may exhibit some effect on the development, timing, and endogenous levels produced? If animals taking in phytoestrogens through the food they eat, or as pharmaceutical products, exhibit effects of the hormones on their own systems, why would plants not have the same effects on them in the uptake of our excreted hormones?
There are currently 6 primary phytohormones (hormones specifically found within plants); auxins, cytokinins, gibberellins (GA), abscisic acid, jasmonats, and ethylene (Kimball, 2006). Another group, the brassinosteroids (BR), is more recently emerging and characterized as influential growth regulators (Halliday, 2004). Generally, all phytohormones play diverse roles in growth and development, and even adaptations to stress. It has even been suggested that plants contain the more mammalian hormones estrone, estradiol, and estriol (Khaleel et al., 2003). Examining the chemical structure of 4 phytohormones, auxin, cytokinin, GA and brassinolind (Figure 5.1), and the structure of mammalian hormones, estrone, estradiol, progesterone and the synthetic estrogen EE2 (Figure 5.1), there are notable similarities. Not only are there structural similarities, but octanol-water coefficients for the hormones are similar, encompassing a range that allows them to diffuse across plasma membranes (Kimball, 2006). BRs are the most structurally similar to mammalian steroids and have been found to influence cell division and elongation of cells, including root cells (Amzallag and Goloubinoff, 2003). Their presence in the roots may allow for an interaction of the structurally similar EE2 with the BR receptor, assuming it has a relatively low degree of specificity.
Hormone signaling in both plants and animals is typically initiated by binding of the ligand to a receptor protein that may be located within the plasma membrane, or is found soluble within the cytosol or nucleus of the cell. Current understanding of plant hormone signaling is limited. Evidence for specific plant hormone receptors is just being documented. In 2005, it was proposed that a soluble GA receptor is present in the nucleus
208 Carmen G. Franks - Appendices
that once GA binds to, induces a chain of events that still remain unclear (Ueguchi- Tanaka et al., 2005). Cytokinins bind to a trans-membrane protein that then induces a chain of signaling (Ferreira and Kieber, 2005). Recent work has linked auxin receptors to a soluble protein (Dharmasiri et al., 2003 and 2005). BR signaling is linked to a membrane bound surface protein (Halliday, 2004; Vert et al., 2005). Research points to the existence of structurally similar animal G-proteins, responsible for transducing the action of a wide variety of extracellular signaling molecules, in plants that may play a role in hormone signaling (Hooley, 1999; Jones and Assmann, 2004). Studies involving Arabidopsis uptake of mammalian estrogens have no problems with the removal of the hormones from the growing medium (Janeczko et al., 2003).
Not only do mammalian hormones affect other animals, but they have been noted to induce effects in plants. When some of these hormones have been applied exogenously to plants cell division was stimulated, pollen germination, and growth and flowering (Geuns, 1978; Bhattacharya and Gupta, 1981; Hayat et al., 2001). As EE2 is a synthetic hormone with its precursor compounds estrone and 17β-estradiol, it could be suggested that EE2 would readily be taken into the plant just as natural hormones are and induce similar effects. In 2003, Janeczko et al. examined the influence of common sex hormones estrone, estriol, 17β-estradiol, androsterone, androstenedione, and progesterone on the flowering induction of Arabidopsis, with the male hormones inducing flowering and the female hormones generally inhibited generative plant numbers. There is some debate that mammalian sex hormones are naturally occurring within plants (Simons and Grinwich, 1989; Milanesi et al., 2001; Khaleel et al., 2003; Milanesi and Boland, 2004).
209 Carmen G. Franks - Appendices
estrone
auxin
estriol
cytokinin
estradiol
abscisic acid
gibberellin
ethynylestradiol
brassinolind progesterone
Figure D1.1. Plant steroid structures: auxin, cytokinin, abscisic acid, gibberellin and brassinolind and mammalian estrogen steroids and progesterone chemical structures: estrone, estriol, estradiol, ethynylestradiol and progesterone.
210 Carmen G. Franks - Appendices
Diltiazem. Diltiazem (DTZ), or commonly known as Cardizem®, is a calcium-channel blocker used as an antihypertensive and muscle relaxant. DTZ is a member of benzothiazepine antihypertensive drugs, commonly grouped as a calcium channel blocker or calcium antagonist. Calcium antagonists were discovered in 1963 by chance, with their own drug designation given in 1969. DTZ joined this group when it was introduced in 1975 by Japanese pharmacologists (Yousef et al., 2005).
DTZ is a member of benzothiazepine antihypertensive drugs, commonly grouped as a calcium channel blocker or calcium antagonist. In animals, it is an L-type calcium channel blocker. Evidence suggests that plants contain Ca2+ channels structurally similar to animal L-type channels, likely providing binding sites within plant root plasma membranes for the pharmaceutical DTZ, without inhibiting Ca2+ uptake.
Mobilization of Ca2+ for signaling requires an initiator. The protein NAADP, has been identified as a potent mobilizer (Genazzani et al., 1997). In higher plants, diltiazem was found to inhibit NAADP (nicotinic acid adenine dinucleotide phosphate (a metabolite of NADP)) signaled release of Ca2+ suggesting structural similarities to animal L-type channels (Genazzani et al., 1997; Navazio et al., 2000). In plant root cells, the structurally similar L-type channels were relatively insensitive to DTZ with the influx of Ca2+ into the cytosol not inhibited by DTZ (Plieth, 2005).
These structurally similar L-type Ca2+ channels within plants could provide a site of action for DTZ and may result in a toxic effect for the plant. Diazepam. Diazepam (DZP) is a benzodiazepine used as antianxiety agents, muscle relaxants, sedatives, hypnotics, and sometimes as anticonvulsants in both humans and animals. The specific chemical DZP is sold under the Valium trademark (Engelberg, 1982). DZP method of synthesis was developed in 1961, with market approval of Valium® granted in 1963 and patented in the United States in 1968 (Mune, 1990).
The environmental fate of DZP was investigated in a water/sediment system by Loffler et al. (2005) and was found to sorb to sediment at an elevated level, with persistence of > 365 days as near 100% parent compound. DZP is considered extremely stable in soils and during sewage treatment and relatively stable in surface waters excluding light-induced degradation (Loffler et al., 2005).
Benzodiazepine substances were originally thought to be a synthetic creation until their discovery in plant and animals in the mid-1980’s (Kavvadias et al., 2000). Now endogenous benzodiazepines, including DZP, have been found in almost every organism tested (Unseld et al., 1989). Found to interact primarily with GABA receptors to exhibit their effect on the CNS of higher animals, now these receptors are being found within plants, bacteria and animals (Kinnersley and Lin, 2000; Papadopoulos, 2004).
Invertebrates form the bottom of many food chains therefore any drug that may impact population numbers is of concern. As DZP functions within the nervous system it can be potentially expected to induce effects in any organism with a nervous system, vertebrate
211 Carmen G. Franks - Appendices
and invertebrate alike (Wildmann, 1988). Studies on nematodes have shown it to induce both relaxant and muscle activating responses via GABA receptors (Richmond and Jorgensen, 1999). Although in crayfish no modulation of the receptors was found (Pearstein et al., 1997). Receptors have been found in the CNS, peripheral systems, and in mitochondria of animals and bacteria (Slocinska et al., 2004).
The presence of GABA receptors in plants has been determined, but their location is still unresolved (Kinnersley and Lin, 2000). The association of GABA binding sites with G- proteins lends another receptor avenue, found within the plasma membrane (Jones and Assmann, 2004). Benzodiazepines have been identified within untreated animals and humans, with some speculation that it came from ingested plants (Kavvadias et al., 2000). It could be conceived that these similar receptors in plants could be places for the benzodiazepines to bind, as they do in animals, including DZP.
There is strong evidence that the roles of these receptors include steroid biosynthesis, control of mitochondrial respiration, cell proliferation, flow of calcium ions, cellular immunity, malignancy, and apoptosis (Slocinska et al., 2004). This variety of roles provides rationale for the presence of these compounds in plants. It could be possible for these natural benzodiazepines enhancing properties to assist these processes in plants and particularly in germination (Wildmann, 1988).
In potato, benzodiazepines and benzodiazepine receptors have been identified (Kavvadias et al., 2000; Corsi et al., 2004). It has been suggested that the existence of benzodiazepines and their receptors occurred before the divergence of the kingdoms and has been conserved across species (Papadopoulos, 2004). The exact role of these compounds and receptors in plants is not known, but evidence suggests roles in ion control (Kinnersley and Lin, 2000). In 1988, it was reported that levels of natural benzodiazepines increase by five-fold in seedlings during germination of wheat and potato, than of non-germinating tissue, suggesting a role of these compounds in the development of higher plants (Wildmann, 1988).
DZP has been reported to be oxidized by Cytochrome P450s (CYPs), particularly CYP3A4 to temazepam (hydroxylation) and CYP2C19 and CYP3A4 to N-demethyl- diazepam (N-demethylation) in man (Niwa et al., 2005). CYPs are ubiquitous enzymes capable of catalyzing metabolic reactions in plants and animals. It is possible that similar metabolites could be formed in plants, as in man.
212 Carmen G. Franks - Appendices
Atrazine. Atrazine (ATZ) is one of the mostly widely used broad leaf herbicides (Hayes et al., 2002). ATZ is applied as both a pre- and post-emergence spray primarily in corn, soybean and sorghum. Originally developed in Switzerland in 1952 and patented in 1958, it was registered for commercial use in the United States by 1959 (Solomon et al., 1996). The use of atrazine has since spread to most of the world. Efforts are in effect to reduce the quantity used after environmental concerns were registering.
A member of the s-Triazine family of herbicides, ATZ acts to interrupt photosynthesis resulting in oxidative stress leading to death. ATZ does not ‘starve’ a plant to death, as previously believed. As ATZ binds specifically to certain transfer proteins within the photosynthetic chain, resistance can be developed through mutations in the genes coding for those proteins, or via naturally resistant plants possessing different transfer proteins. Corn, soy and sorghum are some of those naturally resistant plants. In the United States alone, approximately 800 million lb (363 million kg) was applied between 1980 and 1990 (Mandelbaum et al., 1995).
As a positive control, ATZ is known to be taken up by plants, therefore if it was removed from solution during the study then the experiment was set up to work at least for ATZ. Also, ATZ is known to be taken up and metabolized by hybrid poplar trees, of which their physiology could be described as close to willow physiology (they are from the same family Salicaceae), providing a basis for what could be occurring in willow (Burken and Schnoor, 1997 and 1998).
Water is the primary dispersing force for atrazine. Due to varying solubility, the parent compound is commonly found within runoff, while some degradation products such as deethylatrazine and deisopropylatrazine are more soluble and commonly found within ground water (Angier et al., 2002).
Reports have provided differing rates of degradation in both aerobic and anaerobic environments for atrazine. Seybold et al. (2001) reported a half-life of 38 days in aerobic wetland soil, and 86 days in the aqueous phase above the soil. Chung et al. (1995), also in anaerobic wetland soils, reported a half-life of 38 weeks (266 days). Other studies in an anaerobic aquatic environment, reported overall atrazine half-life of 608 days, a water half-life of 578 days, and in sediment 330 days (EPA, 2002).
213 Carmen G. Franks - Appendices
LITERATURE CITED
Amzallag, GN and Goloubinoff, P. 2003. An Hsp90 inhibitor, geldanamycin, as a brassinosteroid antagonist: Evidence from salt-exposed roots of Vigna radiate. Plant Biol. 5: 143-150. Angier, JT, McCarty, GW, Rice, CP, Bialek, K. 2002. Influence of a riparian wetland on nitrate and herbicides exported from an agricultural field. J. Agric. Food Chem. 50: 4424-4429. Bhattacharya, B and Gupta, K. 1981. Steroid hormone effects on growth and apical dominance of sunflower. Phytochemistry 20: 989-991. Burken, JG and Schnoor, JL. 1997. Uptake and metabolism of atrazine by poplar trees. Environ. Sci. Technol. 31: 1399-1406. Burken, JG, and Schnoor, JL. 1998. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32: 3379-3385. Chung, KH, Ro, KS, Roy, D. 1995. Atrazine biotransformation in wetland sediment under different nutrient conditions .1. Anaerobic. J. Environ. Sci. Health A-Environ. Sci. Eng. Toxic Hazard. Subt. Control 30: 109-120. Corsi, L, Avallone, R, Geminiani, E, Cosenza, F, Venturini, I, Baraldi, M. 2004. Peripheral benzodiazepine receptors in potatoes (Solanum tuberosum). Biochem. Biophys. Res. Commun. 313: 62-66. Dharmasiri, N, Dharmasiri, S, Estelle, M. 2005. The F-box protein TIR1 is an auxin receptor. Nature 435: 441-445. Dharmasiri, N, Dharmasiri, S, Jones, AM, Estelle, M. 2003. Auxin action in a cell-free system. Curr. Biol. 13: 1418-1422. Engelberg, AB. 1982. Patent term extension: an overarching solution to a nonexistent problem. Health Aff. (Millwood). Spring, 1: 34-35. EPA. 2002. Reregistration Eligibility Science Chapter for Atrazine Environmental Fate and Effects Chapter, 2002. The U.S. Environmental Protection Agency. Online source: http://www.epa.gov/oppsrrd1/reregistration/atrazine/ Ferreira, FJ and Kieber, JJ. 2005. Cytokinin signaling. Curr. Opin. Plant Biol. 8: 518- 525. Genazzani, AA, Mezna, M, Dickey, DM, Michelangeli, F, Walseth, TF, Galione, A. 1997. Pharmacological properties of the Ca2+-release mechanism sensitive to NAADP in the sea urchin egg. Br. J. Pharmacol. 121: 1489-1495. Geuns, JMC. 1978. Steroid hormones and plant growth and development. Phytochemistry 17: 1-14. Halliday, KJ. 2004. Plant hormones: the interplay of brassinosteroids and auxin. Curr. Biol. Dispatch 14: R1008-R1010. Hayat, S, Ahmad, A, Hussain, A, Mobin, M. 2001. Growth of wheat seedlings raised from the grains treated with 28-homobrassinolide. Acta Physiol. Plant. 23: 27-30. Hayes, TB, Collins, A, Lee, M, Mendoza, M, Noriega, N, Stuart, AA, Vonk, A. 2002. Hermaphroditic, demasculinized frogs after exposure to herbicide atrazine at low ecologically relevant doses. Proc. Natl. Acad. Sci. U. S. A. 99: 5476-5480. Hooley, R. 1999. A role for G proteins in plant hormone signaling? Plant Physiol. Biochem. 37: 393-402.
214 Carmen G. Franks - Appendices
Janeczko, A, Filek, W, Biesaga-Kościelniak, J, Marcińska, I, Janeczko, Z. 2003. The influence of animal sex hormones on the induction of flowering in Arabidopsis thaliana: comparison with the effect of 24-epibrassinolide. Plant Cell Tissue Organ Cult. 72: 147-151. Jones, AM and Assmann, SM. 2004. Plants: the latest model system for G-protein research. EMBO reports 5: 572-578. Kavvadias, D, Abou-Mandour, AA, Czygan, FC, Beckmann, H, Sand, P, Riederer, P, Schreier, P. 2000. Identification of benzodiazepines in Artemisia dracunculus and Solanum tuberosum rationalizing their endogenous formation in plant tissue. Biochem. Biophys. Res. Commun. 269: 290-295. Khaleel, TF, Dillman, R, Gretch, D. 2003. Estradiol distribution during the development and expression of reproductive structures in Populus tremuloides Michx. Sex. Plant Reprod. 16: 35-42. Kimball, JW. 2006. Kimball’s Biology Pages. Online source: http://users.rcn.com/jkimball.ma.ultranet/BiologyPages/ Kinnersley, AM and Lin, F. 2000. Receptor modifiers indicate that 4-aminobutyric acid (GABA) is a potential modulator of ion transport in plants. Plant Growth Regul. 32: 65-76. Loffler, D, Rombke, J, Meller, M, Ternes, T. 2005. Environmental fate of pharmaceuticals in water/sediment systems. Environ. Sci. Technol. 39 (14): 5209- 5218. Mandelbaum, RT, Allan, DL, Wackett, LP. 1995. Isolation and characterization of a Pseudomonas sp. that mineralizes the s-triazine herbicide atrazine. Appl. Environ. Microbiol. 61: 1451-1457. Milanesi, L and Boland, R. 2004. Presence of estrogen receptor (ER)-like proteins and endogenous ligands for ER in solanaceae. Plant Sci. 166: 397-404. Milanesi, L, Monje, P, Boland, R. 2001. Presence of estrogens and estrogen receptor-like proteins in Solanum glaucophyllum. Biochem. Biophys. Res. Commun. 289: 1175- 1179. Moriyama, K, Matsufuji, H, Chino, M, Takeda, M. 2004. Identification and behavior of reaction products formed by chlorination of ethynylestradiol. Chemosphere 55: 839-847. Mune, P. 1990. Diazepam. IPCS INCHEM. Online source: http://www.inchem.org/ Navazio, Lorella, Bewell, MA, Siddiqua, A, Dickinson, GD, Galione, A, Sanders, D. 2000. Calcium release from the endoplasmic reticulum of higher plants elicited by the NASP metabolite nicotinic acid adenine dinucleotide phosphate. Proc. Natl. Acad. Sci. U. S. A. 97: 8693-8698. Niwa, T, Shiraga, T, Ishii, I, Kagayama, A, Takagi, A. 2005. Contribution of human hepatic Cytochrome P450 isoforms to the metabolism of psychotropic drugs. Biol. Pharm. Bull. 28: 1711-1716. Papadopoulos, V. 2004. In search of the function of the peripheral-type benzodiazepine receptor. Endocr. Res. 30: 677-684. Pearstein, E, Cattaert, D, Clarac, F. 1997. Crayfish sensory terminals and motor neurones exhibit two distinct types of GABA receptors. J. Comp. Physiol. A 180: 71-79. Plieth, C. 2005. Calcium: just another regulator in the machinery of life? Ann. Bot. 96: 1- 5.
215 Carmen G. Franks - Appendices
Reed, MJ, Fotherby, K, Steele, SJ. 1972. Metabolism of ethynyloestradiol in man. J. Endocr. 55: 351-361. Richmond, JE, Jorgensen, EM. 1999. One GABA and two acetylcholine receptors function at the C-elegans neuromuscular junction. Nat. Neurosci. 2: 791-797. Seybold, CA, Mersie, W, McNamee, C. 2001. Anaerobic degradation of atrazine and metachlor and metabolite formation in wetland soil and water microcosms. J. Environ. Qual. 30: 1271-1277. Simons, RG and Grinwich, DL. 1989. Immunoreactive detection of four mammalian steroids in plants. Can. J. Bot. 67: 288-296. Slocinska, M, Szewczyk, A, Hryniewiecka, L, Kmita, H. 2004. Benzodiazepine binding to mitochondrial membranes of the amoeba Acanthamoeba castellanii and the yeast Saccharomyces cerevisiae. Acta Biochim. Pol. 51: 953-962. Solomon, K, Baker, DB, Richards, PR, Dixon, KR, Klaine, SJ, La Point, TW, Kendall, RJ, Weisskopf, CP, Giddings, JM, Giesy, JP, Hall, LW Jr, Williams, WM. 1996. Ecological risk assessment of atrazine in North American surface waters. Environ. Toxicol. Chem. 15: 31-76. Ueguchi-Tanaka, M, Ashikari, M, Nakajima, M, Itoh, H, Katoch, E, Kobayashi, M, Chow, T, Hsing, YC, Kitano, H, Yamaguchi, I, Matsuoka, M. 2005. Gibberellin insensitive dwarf1 encodes a soluble receptor for gibberellin. Nature 437 (29): 693- 698. Unseld, E, Kirshna, DR, Fischer, C, Klotz, U. 1989. Detection of desmethyldiazepam and diazepam in brain of different species and plants. Biochem. Pharmacol. 38: 2473- 2478. Vert, G, Nemhauser, J L, Geldner, N, Hong, F, Chory, J. 2005. Molecular mechanisms of steroid hormone signaling in plants. Annu. Rev. Cell Dev. Biol. 21: 177-201. Wildmann, J. 1988. Increase of natural benzodiazepines in wheat and potato during germination. Biochem. Biophys. Res. Commun. 157: 1436-1443. Yousef, W, Omar, M, Adel, H, Dorsy, MD, El-Wahed, MMA, Ghanayem, NM. 2005. The mechanism of action of calcium channel blockers in the treatment of diabetic nephropathy. Int. J. Diabetes & Metabolism 13: 76-82.
216